v13.2.2 Mimic

This is the second bugfix release of the Mimic v13.2.x long term stable release series. This release contains many fixes across all components of Ceph. We recommend that all users upgrade.

Changelog

v13.2.1 Mimic

This is the first bugfix release of the Mimic v13.2.x long term stable release series. This release contains many fixes across all components of Ceph, including a few security fixes. We recommend that all users upgrade.

Notable Changes

  • CVE 2018-1128: auth: cephx authorizer subject to replay attack (issue#24836, Sage Weil)
  • CVE 2018-1129: auth: cephx signature check is weak (issue#24837, Sage Weil)
  • CVE 2018-10861: mon: auth checks not correct for pool ops (issue#24838, Jason Dillaman)

Changelog

v13.2.0 Mimic

This is the first stable release of Mimic, the next long term release series.

Major Changes from Luminous

  • Dashboard:
    • The (read-only) Ceph manager dashboard introduced in Ceph Luminous has been replaced with a new implementation inspired by and derived from the openATTIC Ceph management tool, providing a drop-in replacement offering a number of additional management features.
  • RADOS:
    • Config options can now be centrally stored and managed by the monitor.
    • The monitor daemon uses significantly less disk space when undergoing recovery or rebalancing operations.
    • An async recovery feature reduces the tail latency of requests when the OSDs are recovering from a recent failure.
    • OSD preemption of scrub by conflicting requests reduces tail latency.
  • RGW:
    • RGW can now replicate a zone (or a subset of buckets) to an external cloud storage service like S3.
    • RGW now supports the S3 multi-factor authentication API on versioned buckets.
    • The Beast frontend is no longer experimental, and is considered stable and ready for use.
  • CephFS:
    • Snapshots are now stable when combined with multiple MDS daemons.
  • RBD:
    • Image clones no longer require explicit protect and unprotect steps.
    • Images can be deep-copied (including any clone linkage to a parent image and associated snapshots) to new pools or with altered data layouts.
  • Misc:
    • We have dropped the Debian builds for the Mimic release due to the lack of GCC 8 in Stretch. We expect Debian builds to return with the release of Buster in early 2019, and hope to build a final Luminous release (and possibly later Mimic point releases) once Buster is available.

Upgrading from Luminous

Notes

  • We recommend you avoid creating any RADOS pools while the upgrade is in process.
  • You can monitor the progress of your upgrade at each stage with the ceph versions command, which will tell you what ceph version(s) are running for each type of daemon.

Instructions

  1. If your cluster was originally installed with a version prior to Luminous, ensure that it has completed at least one full scrub of all PGs while running Luminous. Failure to do so will cause your monitor daemons to refuse to join the quorum on start, leaving them non-functional.

    If you are unsure whether or not your Luminous cluster has completed a full scrub of all PGs, you can check your cluster’s state by running:

    # ceph osd dump | grep ^flags
    

    In order to be able to proceed to Mimic, your OSD map must include the recovery_deletes and purged_snapdirs flags.

    If your OSD map does not contain both these flags, you can simply wait for approximately 24-48 hours, which in a standard cluster configuration should be ample time for all your placement groups to be scrubbed at least once, and then repeat the above process to recheck.

    However, if you have just completed an upgrade to Luminous and want to proceed to Mimic in short order, you can force a scrub on all placement groups with a one-line shell command, like:

    # ceph pg dump pgs_brief | cut -d " " -f 1 | xargs -n1 ceph pg scrub
    

    You should take into consideration that this forced scrub may possibly have a negative impact on your Ceph clients’ performance.

  2. Make sure your cluster is stable and healthy (no down or recovering OSDs). (Optional, but recommended.)

  3. Set the noout flag for the duration of the upgrade. (Optional, but recommended.):

    # ceph osd set noout
    
  4. Upgrade monitors by installing the new packages and restarting the monitor daemons.:

    # systemctl restart ceph-mon.target
    

    Once all monitors are up, verify that the monitor upgrade is complete by looking for the mimic feature string in the mon map. For example:

    # ceph mon feature ls
    

    should include mimic under persistent features:

    on current monmap (epoch NNN)
       persistent: [kraken,luminous,mimic]
       required: [kraken,luminous,mimic]
    
  5. Upgrade ceph-mgr daemons by installing the new packages and restarting with:

    # systemctl restart ceph-mgr.target
    

    Verify the ceph-mgr daemons are running by checking ceph -s:

    # ceph -s
    
    ...
      services:
       mon: 3 daemons, quorum foo,bar,baz
       mgr: foo(active), standbys: bar, baz
    ...
    
  6. Upgrade all OSDs by installing the new packages and restarting the ceph-osd daemons on all hosts:

    # systemctl restart ceph-osd.target
    

    You can monitor the progress of the OSD upgrades with the new ceph versions or ceph osd versions command:

    # ceph osd versions
    {
       "ceph version 12.2.5 (...) luminous (stable)": 12,
       "ceph version 13.2.0 (...) mimic (stable)": 22,
    }
    
  7. Upgrade all CephFS MDS daemons. For each CephFS file system,

    1. Reduce the number of ranks to 1. (Make note of the original number of MDS daemons first if you plan to restore it later.):

      # ceph status
      # ceph fs set <fs_name> max_mds 1
      
    2. Wait for the cluster to deactivate any non-zero ranks by periodically checking the status:

      # ceph status
      
    3. Take all standby MDS daemons offline on the appropriate hosts with:

      # systemctl stop ceph-mds@<daemon_name>
      
    4. Confirm that only one MDS is online and is rank 0 for your FS:

      # ceph status
      
    5. Upgrade the last remaining MDS daemon by installing the new packages and restarting the daemon:

      # systemctl restart ceph-mds.target
      
    6. Restart all standby MDS daemons that were taken offline:

      # systemctl start ceph-mds.target
      
    7. Restore the original value of max_mds for the volume:

      # ceph fs set <fs_name> max_mds <original_max_mds>
      
  8. Upgrade all radosgw daemons by upgrading packages and restarting daemons on all hosts:

    # systemctl restart radosgw.target
    
  9. Complete the upgrade by disallowing pre-Mimic OSDs and enabling all new Mimic-only functionality:

    # ceph osd require-osd-release mimic
    
  10. If you set noout at the beginning, be sure to clear it with:

    # ceph osd unset noout
    
  11. Verify the cluster is healthy with ceph health.

Upgrading from pre-Luminous releases (like Jewel)

You must first upgrade to Luminous (12.2.z) before attempting an upgrade to Mimic. In addition, your cluster must have completed at least one scrub of all PGs while running Luminous, setting the recovery_deletes and purged_snapdirs flags in the OSD map.

Upgrade compatibility notes

These changes occurred between the Luminous and Mimic releases.

  • core:

    • The pg force-recovery command will not work for erasure-coded PGs when a Luminous monitor is running along with a Mimic OSD. Please use the recommended upgrade order of monitors before OSDs to avoid this issue.
    • The sample crush-location-hook script has been removed. Its output is equivalent to the built-in default behavior, so it has been replaced with an example in the CRUSH documentation.
    • The -f option of the rados tool now means --format instead of --force, for consistency with the ceph tool.
    • The format of the config diff output via the admin socket has changed. It now reflects the source of each config option (e.g., default, config file, command line) as well as the final (active) value.
    • Commands variously marked as del, delete, remove etc. should now all be normalized as rm. Commands already supporting alternatives to rm remain backward-compatible. This changeset applies to the radosgw-admin tool as well.
    • Monitors will now prune on-disk full maps if the number of maps grows above a certain number (mon_osdmap_full_prune_min, default: 10000), thus preventing unbounded growth of the monitor data store. This feature is enabled by default, and can be disabled by setting mon_osdmap_full_prune_enabled to false.
    • rados list-inconsistent-obj format changes:
      • Various error strings have been improved. For example, the “oi” or “oi_attr” in errors which stands for object info is now “info” (e.g. oi_attr_missing is now info_missing).
      • The object’s “selected_object_info” is now in json format instead of string.
      • The attribute errors (attr_value_mismatch, attr_name_mismatch) only apply to user attributes. Only user attributes are output and have the internal leading underscore stripped.
      • If there are hash information errors (hinfo_missing, hinfo_corrupted, hinfo_inconsistency) then “hashinfo” is added with the json format of the information. If the information is corrupt then “hashinfo” is a string containing the value.
      • If there are snapset errors (snapset_missing, snapset_corrupted, snapset_inconsistency) then “snapset” is added with the json format of the information. If the information is corrupt then “snapset” is a string containing the value.
      • If there are object information errors (info_missing, info_corrupted, obj_size_info_mismatch, object_info_inconsistency) then “object_info” is added with the json format of the information instead of a string. If the information is corrupt then “object_info” is a string containing the value.
    • rados list-inconsistent-snapset format changes:
      • Various error strings have been improved. For example, the “ss_attr” in errors which stands for snapset info is now “snapset” (e.g. ss_attr_missing is now snapset_missing). The error snapset_mismatch has been renamed to snapset_error to better reflect what it means.
      • The head snapset information is output in json format as “snapset.” This means that even when there are no head errors, the head object will be output when any shard has an error. This head object is there to show the snapset that was used in determining errors.
    • The osd_mon_report_interval_min option has been renamed to osd_mon_report_interval, and the osd_mon_report_interval_max (unused) has been eliminated. If this value has been customized on your cluster then your configuration should be adjusted in order to avoid reverting to the default value.
    • The config-key interface can store arbitrary binary blobs but JSON can only express printable strings. If binary blobs are present, the ‘ceph config-key dump’ command will show them as something like <<< binary blob of length N >>>.
    • Bootstrap auth keys will now be generated automatically on a fresh deployment; these keys will also be generated, if missing, during upgrade.
    • The osd force-create-pg command now requires a force option to proceed because the command is dangerous: it declares that data loss is permanent and instructs the cluster to proceed with an empty PG in its place, without making any further efforts to find the missing data.

    CephFS:

    • Upgrading an MDS cluster to 12.2.3+ will result in all active MDS exiting due to feature incompatibilities once an upgraded MDS comes online (even as standby). Operators may ignore the error messages and continue upgrading/restarting or follow this upgrade sequence:

      After upgrading the monitors to Mimic, reduce the number of ranks to 1 (ceph fs set <fs_name> max_mds 1), wait for all other MDS to deactivate, leaving the one active MDS, stop all standbys, upgrade the single active MDS, then upgrade/start standbys. Finally, restore the previous max_mds.

      !! NOTE: see release notes on snapshots in CephFS if you have ever enabled snapshots on your file system.

      See also: https://tracker.ceph.com/issues/23172

    • Several ceph mds ... commands have been obsoleted and replaced by equivalent ceph fs ... commands:

      • mds dump -> fs dump
      • mds getmap -> fs dump
      • mds stop -> mds deactivate
      • mds set_max_mds -> fs set max_mds
      • mds set -> fs set
      • mds cluster_down -> fs set cluster_down true
      • mds cluster_up -> fs set cluster_down false
      • mds add_data_pool -> fs add_data_pool
      • mds remove_data_pool -> fs rm_data_pool
      • mds rm_data_pool -> fs rm_data_pool
    • New CephFS file system attributes session_timeout and session_autoclose are configurable via ceph fs set. The MDS config options mds_session_timeout, mds_session_autoclose, and mds_max_file_size are now obsolete.

    • As the multiple MDS feature is now standard, it is now enabled by default. ceph fs set allow_multimds is now deprecated and will be removed in a future release.

    • As the directory fragmentation feature is now standard, it is now enabled by default. ceph fs set allow_dirfrags is now deprecated and will be removed in a future release.

    • MDS daemons now activate and deactivate based on the value of max_mds. Accordingly, ceph mds deactivate has been deprecated as it is now redundant.

    • Taking a CephFS cluster down is now done by setting the down flag which deactivates all MDS. For example: ceph fs set cephfs down true.

    • Preventing standbys from joining as new actives (formerly the now deprecated cluster_down flag) on a file system is now accomplished by setting the joinable flag. This is useful mostly for testing so that a file system may be quickly brought down and deleted.

    • New CephFS file system attributes session_timeout and session_autoclose are configurable via ceph fs set. The MDS config options mds_session_timeout, mds_session_autoclose, and mds_max_file_size are now obsolete.

    • Each mds rank now maintains a table that tracks open files and their ancestor directories. Recovering MDS can quickly get open files’ paths, significantly reducing the time of loading inodes for open files. MDS creates the table automatically if it does not exist.

    • CephFS snapshot is now stable and enabled by default on new filesystems. To enable snapshot on existing filesystems, use the command:

      ceph fs set <fs_name> allow_new_snaps
      

      The on-disk format of snapshot metadata has changed. The old format metadata can not be properly handled in multiple active MDS configuration. To guarantee all snapshot metadata on existing filesystems get updated, perform the sequence of upgrading the MDS cluster strictly.

      See http://docs.ceph.com/docs/mimic/cephfs/upgrading/

      For filesystems that have ever enabled snapshots, the multiple-active MDS feature is disabled by the mimic monitor daemon. This will cause the “restore previous max_mds” step in above URL to fail. To re-enable the feature, either delete all old snapshots or scrub the whole filesystem:

      • ceph daemon <mds of rank 0> scrub_path / force recursive repair
      • ceph daemon <mds of rank 0> scrub_path '~mdsdir' force recursive repair
    • Support has been added in Mimic for quotas in the Linux kernel client as of v4.17.

      See http://docs.ceph.com/docs/mimic/cephfs/quota/

    • Many fixes have been made to the MDS metadata balancer which distributes load across MDS. It is expected that the automatic balancing should work well for most use-cases. In Luminous, subtree pinning was advised as a manual workaround for poor balancer behavior. This may no longer be necessary so it is recommended to try experimentally disabling pinning as a form of load balancing to see if the built-in balancer adequately works for you. Please report any poor behavior post-upgrade.

    • NFS-Ganesha is an NFS userspace server that can export shares from multiple file systems, including CephFS. Support for this CephFS client has improved significantly in Mimic. In particular, delegations are now supported through the libcephfs library so that Ganesha may issue delegations to its NFS clients allowing for safe write buffering and coherent read caching. Documentation is also now available: http://docs.ceph.com/docs/mimic/cephfs/nfs/

    • MDS uptime is now available in the output of the MDS admin socket status command.

    • MDS performance counters for client requests now include average latency as well as the count.

  • RBD

    • The RBD C API’s rbd_discard method now enforces a maximum length of 2GB to match the C++ API’s Image::discard method. This restriction prevents overflow of the result code.
    • The rbd CLI’s lock list JSON and XML output has changed.
    • The rbd CLI’s showmapped JSON and XML output has changed.
    • RBD now optionally supports simplified image clone semantics where non-protected snapshots can be cloned; and snapshots with linked clones can be removed and the space automatically reclaimed once all remaining linked clones are detached. This feature is enabled by default if the OSD “require-min-compat-client” flag is set to mimic or later; or can be overridden via the “rbd_default_clone_format” configuration option.
    • RBD now supports deep copy of images that preserves snapshot history.
  • RGW

    • The RGW Beast frontend is now declared stable and ready for production use. HTTP Frontends for details.
    • Civetweb frontend has been updated to the latest 1.10 release.
    • The S3 API now has support for multi-factor authentication. Refer to RGW Support for Multifactor Authentication for details.
    • RGW now has a sync plugin to sync to AWS and clouds with S3-like APIs.
  • MGR

    • The (read-only) Ceph manager dashboard introduced in Ceph Luminous has been replaced with a new implementation, providing a drop-in replacement offering a number of additional management features. To access the new dashboard, you first need to define a username and password and create an SSL certificate. See the Ceph Manager Dashboard for a feature overview and installation instructions.

    • The ceph-rest-api command-line tool (obsoleted by the MGR restful module and deprecated since v12.2.5) has been dropped.

      There is a MGR module called restful which provides similar functionality via a “pass through” method. See http://docs.ceph.com/docs/master/mgr/restful for details.

    • New command to track throughput and IOPS statistics, also available in ceph -s and previously in ceph -w. To use this command, enable the iostat Manager module and invoke it using ceph iostat. See the iostat documentation for details.

  • build/packaging

    • The rcceph script (systemd/ceph in the source code tree, shipped as /usr/sbin/rcceph in the ceph-base package for CentOS and SUSE) has been dropped. This script was used to perform admin operations (start, stop, restart, etc.) on all OSD and/or MON daemons running on a given machine. This functionality is provided by the systemd target units (ceph-osd.target, ceph-mon.target, etc.).
    • The python-ceph-compat package is declared deprecated, and will be dropped when all supported distros have completed the move to Python 3. It has already been dropped from those supported distros where Python 3 is standard and Python 2 is optional (currently only SUSE).
    • Ceph codebase has now moved to the C++-17 standard.
    • The Ceph LZ4 compression plugin is now enabled by default, and introduces a new build dependency.

Detailed Changelog

  • arch/arm: set ceph_arch_aarch64_crc32 only if the build host supports crc32cx (issue#19705, pr#17420, Kefu Chai)
  • assert(false)->ceph_abort() (pr#18072, Li Wang)
  • auth: keep /dev/urandom open for get_random_bytes (issue#21401, pr#17972, Casey Bodley)
  • bluestore: BlueStore::ExtentMap::dup impl (pr#19719, Shinobu Kinjo)
  • bluestore: bluestore/NVMEDevice: accurate the latency perf counter of queue latency (pr#17435, Ziye Yang, Pan Liu)
  • bluestore: bluestore/NVMEDevice: convert the legacy config opt related with SPDK (pr#18502, Ziye Yang)
  • bluestore: bluestore/NVMEDevice: do not deference a dangling pointer (pr#19067, Kefu Chai)
  • bluestore: bluestore/NVMEDevice: fix the bug in write function (pr#17086, Ziye Yang, Pan Liu)
  • bluestore: bluestore/NVMeDevice: update NVMeDevice code due to SPDK upgrade (pr#16927, Ziye Yang)
  • bluestore,build/ops: bluestore,cmake: enable building bluestore without aio (pr#19017, Kefu Chai)
  • bluestore,build/ops: Build: create a proper WITH_BLUESTORE option (pr#18357, Alan Somers)
  • bluestore,build/ops: ceph.spec.in,debian/rules: change aio-max-nr to 1048576 (pr#17894, chenliuzhong)
  • bluestore,build/ops,tests: os: add compile option to build libbluefs.so (pr#16733, Pan Liu)
  • bluestore,build/ops,tests: test/fio: fix build failure caused by sequencer replacement (pr#20387, Igor Fedotov)
  • bluestore: ceph-bluestore-tool: better fsck/repair, bluefs-bdev-{expand,sizes} (pr#17709, Sage Weil)
  • bluestore: ceph-bluestore-tool: check if bdev is empty on ‘bluefs-bdev-expand’ (pr#17874, WANG Guoqin)
  • bluestore: ceph-bluestore-tool: link target shouldn’t ending with “n” (pr#18585, Yao Zongyou)
  • bluestore,common: intarith: get rid of P2* and ROUND_UP* macros (pr#21085, xie xingguo)
  • bluestore: comp_min_blob_size init error (pr#18318, linbing)
  • bluestore: config: Change bluestore_cache_kv_max to type INT64 (pr#20255, Zhi Zhang)
  • bluestore,core: ceph-bluestore-tool: prime-osd-dir: update symlinks instead of bailing (pr#18565, Sage Weil)
  • bluestore,core: common/options: bluefs_buffered_io=true by default (pr#20542, Sage Weil)
  • bluestore,core: os/bluestore: compensate for bad freelistmanager size/blocks metadata (issue#21089, pr#17268, Sage Weil)
  • bluestore,core: os/bluestore: fix data read error injection in bluestore (pr#19866, Sage Weil)
  • bluestore,core: os/bluestore: kv_max -> kv_min (pr#20544, Sage Weil)
  • bluestore,core: os/bluestore: switch default allocator to stupid; test both bitmap and stupid in qa (pr#16906, Sage Weil)
  • bluestore,core: src/bluestore/NVMEDevice: make all read use aio_submit (pr#17655, Ziye Yang, Pan Liu)
  • bluestore,core,tests: test/unittest_bluefs: check whether rmdir success (pr#15363, shiqi)
  • bluestore,core: tool: ceph-kvstore-tool doesn’t umount BlueStore properly (issue#21625, pr#18083, Chang Liu)
  • bluestore: define default value of LoglevelV only once (3 templates) (pr#20727, Matt Benjamin)
  • bluestore: drop unused friend class in SharedDriverQueueData (pr#16894, Pan Liu)
  • bluestore: fix aio_t::rval type (issue#23527, pr#21136, kungf)
  • bluestore: fix build on armhf (pr#20951, Kefu Chai)
  • bluestore: fixed compilation error when enable spdk with gcc 4.8.5 (pr#16945, Ziye Yang, Pan Liu)
  • bluestore: kv/RocksDBStore: extract common code to a new function (pr#16532, Pan Liu)
  • bluestore/NVMEDevice: code cleanup (pr#17284, Ziye Yang, Pan Liu)
  • bluestore: os/bluestore: add bluestore_prefer_deferred_size_hdd/ssd to tracked keys (pr#17459, xie xingguo)
  • bluestore: os/bluestore: add discard method for ssd’s performance (pr#14727, Taeksang Kim)
  • bluestore: os/bluestore: Add lat record of deferred_queued and deferred_aio_wait (pr#17015, lisali)
  • bluestore: os/bluestore: Add missing __func__ in dout (pr#17903, lisali)
  • bluestore: os/bluestore: add perf counter for allocator fragmentation (pr#21377, Igor Fedotov)
  • bluestore: os/bluestore: allocate entire write in one go (pr#17698, Sage Weil)
  • bluestore: os/bluestore: allow reconstruction of osd data dir from bluestore bdev label (pr#18256, Sage Weil)
  • bluestore: os/bluestore: alter the allow_eio policy regarding kernel’s error list (issue#23333, pr#21306, Radoslaw Zarzynski)
  • bluestore: os/bluestore: avoid excessive ops in _txc_release_alloc (pr#18854, Igor Fedotov)
  • bluestore: os/bluestore: avoid omit cache for remove-collection (pr#18785, Jianpeng Ma)
  • bluestore: os/bluestore: avoid overhead of std::function in blob_t (pr#20294, Radoslaw Zarzynski)
  • bluestore: os/bluestore: avoid unneeded BlobRefing in _do_read() (pr#19864, Radoslaw Zarzynski)
  • bluestore: os/bluestore: be more verbose when hitting unloaded shard in extent map (pr#21245, Igor Fedotov)
  • bluestore: os/bluestore/BlueFS: compact log even when sync_metadata sees no work (pr#17354, Sage Weil)
  • bluestore: os/bluestore/BlueFS: Don’t call debug related code under any condition (pr#17627, Jianpeng Ma)
  • bluestore: os/bluestore/BlueFS: don’t need wait for aio when using _sync_write (pr#16066, Haodong Tang)
  • bluestore: os/bluestore/BlueFS: fix race with log flush during async log compaction (issue#21878, pr#18428, Sage Weil)
  • bluestore: os/bluestore/BlueFS: move release unused extents work in _flush_and_syn_log (pr#17684, Jianpeng Ma)
  • bluestore: os/bluestore/BlueFS: prevent _compact_log_async reentry (issue#21250, pr#17503, Sage Weil)
  • bluestore: os/bluestore/BlueFS: Reduce unnecessary operations in collect_metadata (pr#17995, Luo Kexue)
  • bluestore: os/bluestore/BlueFS: sanity check that alloc->allocate() won’t return 0 (pr#18259, xie xingguo)
  • bluestore: os/bluestore/BlueFS: several cleanups (pr#17966, xie xingguo)
  • bluestore: os/bluestore/bluefs_types: make block_mask 64-bit (pr#21629, Sage Weil)
  • bluestore: os/bluestore/BlueStore: ASAP wake up _kv_finalize_thread (pr#18203, Jianpeng Ma)
  • bluestore: os/bluestore/BlueStore: narrow deferred_lock in _deferred_submit_unlock (pr#17628, Jianpeng Ma)
  • bluestore: os/bluestore: bluestore repair should use interval_set::union_insert (pr#20900, Igor Fedotov)
  • bluestore: os/bluestore: cleanup around ExtentList, AllocExtent and bluestore_extent_t classes (pr#20360, Igor Fedotov)
  • bluestore: os/bluestore: clearer comments, not slower code (pr#16872, Mark Nelson)
  • bluestore: os/bluestore: correctly check all block devices to decide if journal is_rotational (issue#23141, pr#20602, Greg Farnum)
  • bluestore: os/bluestore: delete redundant header file in KernelDevice.cc (pr#18631, Jing Li)
  • bluestore: os/bluestore: do not assert if BlueFS rebalance is unable to allocate sufficient space (pr#18494, Igor Fedotov)
  • bluestore: os/bluestore: do not core dump when BlueRocksEnv gets EEXIST error (issue#20871, pr#17357, liuchang0812)
  • bluestore: os/bluestore: do not core dump when we try to open kvstore twice (pr#18161, Chang Liu)
  • bluestore: os/bluestore: do not release empty bluefs_extents_reclaiming (pr#18671, Igor Fedotov)
  • bluestore: os/bluestore: do not segv on kraken upgrade debug print (issue#20977, pr#16992, Sage Weil)
  • bluestore: os/bluestore: don’t re-initialize csum-setting for existing blobs (issue#21175, pr#17398, xie xingguo)
  • bluestore: os/bluestore: do SSD discard on mkfs (pr#20897, Igor Fedotov)
  • bluestore: os/bluestore: drop deferred_submit_lock, fix aio leak (issue#21171, pr#17352, Sage Weil)
  • bluestore: os/bluestore: drop unused function declaration (pr#18075, Li Wang)
  • bluestore: os/bluestore: drop unused param “what” in apply() (pr#17251, songweibin)
  • bluestore: os/bluestore: _dump_onode() don’t prolongate Onode anymore (pr#19841, Radoslaw Zarzynski)
  • bluestore: os/bluestore: dynamic CF configuration; put pglog omap in separate CF (pr#18224, Sage Weil)
  • bluestore: os/bluestore: enlarege aligned_size avoid too many vector(> IOV_MAX) (issue#21932, pr#18828, Jianpeng Ma)
  • bluestore: os/bluestore: ExtentMap::reshard - fix wrong shard length (pr#17334, chenliuzhong)
  • bluestore: os/bluestore: fail early on very large objects (issue#20923, pr#16924, Sage Weil)
  • bluestore: os/bluestore: fix another aio stall/deadlock (issue#21470, pr#18118, Sage Weil)
  • bluestore: os/bluestore: fix broken cap in _balance_bluefs_freespace() (pr#21097, Igor Fedotov)
  • bluestore: os/bluestore: fix clone dirty_range again (issue#20983, pr#16994, Sage Weil)
  • bluestore: os/bluestore: fix dirty_shard off-by-one (pr#16850, Sage Weil)
  • bluestore: os/bluestore: fix exceeding the max IO queue depth in KernelDevice (issue#23246, pr#20996, Radoslaw Zarzynski)
  • bluestore: os/bluestore: fix potential assert when splitting collection (pr#19519, Igor Fedotov)
  • bluestore: os/bluestore: fix SharedBlob unregistration (issue#22039, pr#18805, Sage Weil)
  • bluestore: os/bluestore: fix some code formatting (pr#21037, Gu Zhongyan)
  • bluestore: os/bluestore: fix the allocate in bluefs (pr#19030, tangwenjun)
  • bluestore: os/bluestore: fix the demotion in StupidAllocator::init_rm_free (pr#20430, Kefu Chai)
  • bluestore: os/bluestore: fix the wrong usage for map_any (pr#18939, Jianpeng Ma)
  • bluestore: os/bluestore: fix wrong usage for BlueFS::_allocate (pr#20708, Jianpeng Ma)
  • bluestore: os/bluestore: free the spdk qpair resource correctly in destructor of SharedDriverQueueData (pr#20929, Jianyu Li)
  • bluestore: os/bluestore: handle small main device properly (pr#17416, xie xingguo)
  • bluestore: os/bluestore: ignore 0x2000~2000 extent oddity from luminous upgrade (issue#21408, pr#17845, Sage Weil)
  • bluestore: os/bluestore: implement BlueStore repair (pr#19843, Igor Fedotov)
  • bluestore: os/bluestore: make bluefs behave better near enospc (pr#18120, Sage Weil)
  • bluestore: os/bluestore: mark derivatives of AioContext as final (pr#20227, Radoslaw Zarzynski)
  • bluestore: os/bluestore: move aio_callback{,_priv} to base class BlockDevice (pr#17002, mychoxin)
  • bluestore: os/bluestore: move assert of read/write to base class (pr#17033, mychoxin)
  • bluestore: os/bluestore: move size and block_size to the base class BlockDevice (pr#16886, Pan Liu)
  • bluestore: os/bluestore: no need to fsync when failed to write label (pr#20092, tangwenjun)
  • bluestore: os/bluestore: no trim debug noise if there is no trimming to be done (pr#20684, Sage Weil)
  • bluestore: os/bluestore/NVMEDevice: change write_bl to bl (pr#17145, Ziye Yang, Pan Liu)
  • bluestore: os/bluestore/NVMEDevice: fix the nvme queue depth issue (pr#17200, Ziye Yang, Pan Liu)
  • bluestore: os/bluestore/NVMEDevice: Remove using dpdk thread (pr#17769, Ziye Yang, Pan Liu)
  • bluestore: os/bluestore: OpSequencer: reduce kv_submitted_waiters if _is_all_kv_submitted() return true (pr#18622, Jianpeng Ma)
  • bluestore: os/bluestore: optimize _collection_list (pr#18777, Jianpeng Ma)
  • bluestore: os/bluestore: pass strict flag to bluestore_blob_use_tracker_t::equal() (pr#15705, xie xingguo)
  • bluestore: os/bluestore: Prealloc memory avoid realloc in list_collection (pr#18804, Jianpeng Ma)
  • bluestore: os/bluestore: prevent mount if osd_max_object_size >= 4G (pr#19043, Sage Weil)
  • bluestore: os/bluestore: print aio in batch (pr#18873, Kefu Chai)
  • bluestore: os/bluestore: print leaked extents to debug output (pr#17225, Sage Weil)
  • bluestore: os/bluestore: propagate read-EIO to high level callers (pr#17744, xie xingguo)
  • bluestore: os/bluestore: put cached attrs in correct mempool (issue#21417, pr#18001, Sage Weil)
  • bluestore: os/bluestore: recalc_allocated() when decoding bluefs_fnode_t (issue#23212, pr#20701, Jianpeng Ma, Kefu Chai)
  • bluestore: os/bluestore: reduce meaningless flush (pr#19027, tangwenjun)
  • bluestore: os/bluestore: refactor FreeListManager to get clearer view on the number (issue#22535, pr#19718, Igor Fedotov)
  • bluestore: os/bluestore: release disk extents in bulky manner (pr#17913, Radoslaw Zarzynski)
  • bluestore: os/bluestore: remove ineffective BlueFS fnode extent calculation (pr#18905, Igor Fedotov)
  • bluestore: os/bluestore: remove unused parameters (pr#18635, Jianpeng Ma)
  • bluestore: os/bluestore: remove unused variable (pr#21063, Gu Zhongyan)
  • bluestore: os/bluestore: remove useless function submit (pr#17537, mychoxin)
  • bluestore: os/bluestore: reorder members of bluefs_extent_t for space efficiency (pr#21034, Radoslaw Zarzynski)
  • bluestore: os/bluestore: replace dout with ldout in StupidAllocator (pr#17404, Radoslaw Zarzynski)
  • bluestore: os/bluestore: report error and quit correctly when disk error happens (issue#21263, pr#17522, Pan Liu)
  • bluestore: os/bluestore: Revert “os/bluestore: allow multiple DeferredBatches in flight at once” (issue#20925, issue#20295, pr#16900, Sage Weil)
  • bluestore: os/bluestore: s/bluefs_total/bluefs_free/ (pr#21036, xie xingguo)
  • bluestore: os/bluestore: separate finisher for deferred_try_submit (issue#21207, pr#17409, Sage Weil)
  • bluestore: os/bluestore: set bitmap freelist resolution to min_alloc_size (pr#17610, Sage Weil)
  • bluestore: os/bluestore: shrink aio submit size to pending value (pr#17588, kungf)
  • bluestore: os/bluestore: silence -Wreturn-type warning (pr#18286, Kefu Chai)
  • bluestore: os/bluestore: support calculate cost when using spdk (pr#17091, Ziye Yang, Pan Liu)
  • bluestore: os/bluestore: synchronous on_applied completions (pr#18196, Sage Weil)
  • bluestore: os/bluestore: trim cache every 50ms (instead of 200ms) (pr#20498, Sage Weil)
  • bluestore: os/bluestore: update description for bluestore_compression_[min|max|_blob_size options (pr#21244, Igor Fedotov)
  • bluestore: os/bluestore: using macro OBJECT_MAX_SIZE to check osd_max_object_size (pr#19622, Jianpeng Ma)
  • bluestore: osd/bluestore: delete unused variable in KernelDevice (pr#20857, Leo Zhang)
  • bluestore: osd,os/bluestore: Display current size of osd_max_object_size (pr#19725, Shinobu Kinjo)
  • bluestore: Revert “os/bluestore: pass strict flag to bluestore_blob_use_tracker_t::equal()” (issue#21293, pr#17569, Sage Weil)
  • bluestore,rgw: rgw,unittest_bit_alloc: silence clang analyzer warning (pr#17294, Kefu Chai)
  • bluestore,tests: objectstore/store_test: fix lack of flush prior to collection_empty()… (issue#22409, pr#19764, Igor Fedotov)
  • bluestore,tests: Revert “bluestore/fio: Fixed problem with all objects having the same hash (pr#18352, Radoslaw Zarzynski)
  • bluestore,tools: ceph-bluestore-tool: create out_dir before create full path of kvdb (pr#18367, Leo Zhang)
  • bluestore,tools: os/bluestore/bluestore_tool: add log-dump command to dump bluefs’s log (pr#18535, Yang Honggang)
  • build: fix dpdk build error (pr#18087, chunmei)
  • build mimic-dev1 with gcc 7 (issue#22438, pr#19548, Kefu Chai)
  • build/ops: automake: remove files required by automake (pr#17937, Kefu Chai)
  • build/ops: blkin: link against lttng-ust-fork (pr#17673, Mohamad Gebai)
  • build/ops: boost: remove boost submodule (pr#17405, Kefu Chai)
  • build/ops: build: do_cmake: allow ARGS to be overridden (pr#19876, Abhishek Lekshmanan)
  • build/ops: build: remove PGMap.cc from libcommon (pr#18496, Sage Weil)
  • build/ops: ceph-disk activate unlocks bluestore data partition (issue#20488, pr#16357, Felix Winterhalter)
  • build/ops: ceph_disk: allow “no fsid” on activate (pr#18991, Dan Mick)
  • build/ops,cephfs: ceph-object-corpus: update to fix make check (pr#21261, Patrick Donnelly)
  • build/ops,cephfs: cmake, test/fs, client: fix build with clang (pr#20392, Adam C. Emerson)
  • build/ops: ceph.spec: use devtoolset-6-gcc-c++ on aarch64 (issue#22301, pr#19341, Kefu Chai)
  • build/ops: ceph-volume: Require lvm2, move to osd package (issue#22443, pr#19529, Theofilos Mouratidis)
  • build/ops: ceph-volume: tests add tests for the is_mounted utility (pr#16962, Alfredo Deza)
  • build/ops: change WITH_SYSTEMD default to ON (pr#20404, Nathan Cutler)
  • build/ops: cmake/BuildBoost: fixes to ready seastar (pr#20616, Kefu Chai, Casey Bodley)
  • build/ops: cmake,deb: install system units using cmake (pr#20618, Kefu Chai)
  • build/ops: cmake: link libcommon with libstdc++ statically if WITH_STATIC_LIBSTDCXX (issue#22438, pr#19515, Kefu Chai)
  • build/ops: cmake,make-dist: bump up boost version to 1.67 (pr#21572, Kefu Chai)
  • build/ops: cmake,mds: detect std::map::merge() before using it (pr#21211, Willem Jan Withagen, Kefu Chai)
  • build/ops: cmake/mgr: use Python 3 virtualenv if mgr subinterpreter is Python 3 (pr#21446, Nathan Cutler)
  • build/ops,common: cmake, common: silence cmake and gcc warnings (issue#23774, pr#21484, Kefu Chai)
  • build/ops: common/time: add time.h for Alpine build (pr#19863, huanwen ren)
  • build/ops,common: Update C++ standard to 14 and clean up (pr#19490, Adam C. Emerson)
  • build/ops,core: ceph-crush-location: remove (pr#19881, Sage Weil)
  • build/ops,core: ceph-volume: do not use –key during mkfs (issue#22283, pr#19276, Kefu Chai, Sage Weil)
  • build/ops,core: /etc/sysconfig/ceph: remove jemalloc option (issue#20557, pr#18487, Sage Weil)
  • build/ops,core: mimic: cmake,common,filestore: silence gcc-8 warnings/errors (pr#21862, Kefu Chai)
  • build/ops,core: mimic: cmake: do not check for aligned_alloc() anymore (issue#23653, pr#22048, Kefu Chai)
  • build/ops,core: msg/async: update to work with dpdk shipped with spdk v17.10 (pr#19470, Kefu Chai)
  • build/ops,core: zstd: Upgrade to v1.3.2 (pr#18407, Adam C. Emerson)
  • build/ops: debian/control: adjust ceph-{osdomap,kvstore,monstore}-tool feature move (issue#22319, pr#19328, Sage Weil)
  • build/ops: debian/control: adjust ceph-{osdomap,kvstore,monstore}-tool feature move (issue#22319, pr#19395, Kefu Chai, Sage Weil)
  • build/ops: debian/control: adjust ceph-{osdomap,kvstore,monstore}-tool feature move (pr#19356, Kefu Chai)
  • build/ops: debian: fix package relationships after 40caf6a6 (issue#21762, pr#18474, Kefu Chai)
  • build/ops: debian: lock ceph user during purge (pr#15118, Caleb Boylan)
  • build/ops: debian/rules: no more ChangeLog (pr#18023, Sage Weil)
  • build/ops: debian/rules: strip ceph-base libraries (issue#22640, pr#19870, Sage Weil)
  • build/ops: do_{cmake,freebsd}: Don’t invoke nproc(1) on FreeBSD (pr#17949, Alan Somers)
  • build/ops: dpdk: remove redundant dpdk submodule (pr#18712, chunmei)
  • build/ops: EventKqueue: Clang want realloc return to be typed (pr#21550, Willem Jan Withagen)
  • build/ops: filestore,rgw: fix types/casts making clang on 32-Bit working (pr#21055, Daniel Glaser)
  • build/ops: Fix ppc64 support for ceph (pr#16753, Boris Ranto)
  • build/ops: Fix two dpdk assert happened in dpdk library (pr#18409, chunmei)
  • build/ops: FreeBSD: add new required packages to be installed (pr#21349, Willem Jan Withagen)
  • build/ops: githubmap: add some known Ceph reviewers (pr#17507, Patrick Donnelly)
  • build/ops: .githubmap: Add wjwithagen as a known Ceph reviewer (pr#17518, Willem Jan Withagen)
  • build/ops: .githubmap: Update (pr#18230, Sage Weil)
  • build/ops: .gitignore: allow debian .patch files (pr#17577, Ken Dreyer)
  • build/ops: include: compat.h, fix the return result of pthread_set_name() (pr#20474, Willem Jan Withagen)
  • build/ops: install-deps: Add support for ‘opensuse-tumbleweed’ (pr#21650, Ricardo Marques)
  • build/ops: install-deps.sh: avoid re-installing g++-7 (pr#19468, Kefu Chai)
  • build/ops: install-deps.sh, cmake: use GCC-7 on xenial also (pr#19418, Kefu Chai)
  • build/ops: install-deps.sh: install new gcc as the default the right way (pr#19417, Kefu Chai)
  • build/ops: install-deps.sh: pass –no-recommends to zypper (issue#22998, pr#20434, Nathan Cutler)
  • build/ops: install-deps.sh: set python2 %bcond by environment (issue#22999, pr#20436, Nathan Cutler)
  • build/ops: install-deps.sh: use DTS on centos if GCC is too old (pr#19398, Kefu Chai)
  • build/ops: install-deps.sh: use tee for writing a file (pr#19516, Kefu Chai)
  • build/ops: install-deps: use DTS-7 on aarch64 and only download mirrored package indexes (pr#19645, Kefu Chai, Songbo Wang)
  • build/ops: libmpem: Revert “submodule: make libmpem as a submodule.” (pr#18414, Jianpeng Ma)
  • build/ops: logrotate: add systemd reload in logrotate in case of centos minimal without killall (pr#16586, Tianshan Qu)
  • build/ops: make-dist,cmake: avoid re-downloading boost (pr#19124, Kefu Chai)
  • build/ops: make-dist,cmake: move boost tarball location to download.ceph.com (pr#17980, Sage Weil)
  • build/ops: make-dist,cmake: Try multiple URLs to download boost before failing (pr#18048, Brad Hubbard)
  • build/ops: make-dist: fall back to python3 (pr#21127, Nathan Cutler)
  • build/ops,mgr: mgr/dashboard: build tweaks (pr#20752, John Spray)
  • build/ops,mgr: mgr/dashboard: remove node/npm system installation (pr#20898, Tiago Melo)
  • build/ops,mgr: packaging: explicit jinja2 dependency for dashboard (issue#22457, pr#19598, John Spray)
  • build/ops,mgr,tests: mgr/dashboard: replace dashboard with dashboard_v2 (pr#20912, Ricardo Dias)
  • build/ops: mimic: cmake: use javac -h for creating JNI native headers (issue#24012, pr#21824, Kefu Chai)
  • build/ops: mimic: silence various warnings to enable GCC-8 build (pr#22081, Adam C. Emerson, Kefu Chai)
  • build/ops: mon,osd: do not use crush_device_class file to initalize class for new osds (pr#19939, Sage Weil)
  • build/ops: mstart.sh: support read CLUSTERS_LIST from env var (pr#16988, Jiaying Ren)
  • build/ops: os/CMakeLists: fix link errro when enable WITH_PMEM=ON (pr#20658, Jianpeng Ma)
  • build/ops: osdc,os,osd: fix build on osx (pr#20029, Kefu Chai)
  • build/ops: python-numpy-devel build dependency for SUSE (issue#21176, pr#17366, Nathan Cutler)
  • build/ops: qa/tests - added for the suites with subset be able to use ‘testing’ … (pr#21454, Yuri Weinstein)
  • build/ops,rbd: ceph-dencoder: moved RBD types outside of RGW preprocessor guard (issue#22321, pr#19343, Jason Dillaman)
  • build/ops: rbdmap: fix umount when multiple mounts use the same RBD (pr#17978, Alexandre Marangone)
  • build/ops: Revert “make-dist: add OBS-specific release suffix on SUSE” (pr#20813, Nathan Cutler)
  • build/ops,rgw: radosgw: Make compilation with CryptoPP possible (pr#14955, Adam Kupczyk)
  • build/ops: rocksdb: do not use aligned_alloc (issue#23653, pr#21632, Kefu Chai)
  • build/ops: rpm: adjust ceph-{osdomap,kvstore,monstore}-tool feature move (issue#22558, pr#19777, Kefu Chai)
  • build/ops: rpm: build-depends on “cunit-devel” for suse (pr#18997, Kefu Chai)
  • build/ops: rpm: conditionalize Python 2 availability to enable Ceph build on Python 3-only system (pr#20018, Nathan Cutler)
  • build/ops: rpm,debian: Ensure all ceph-disk runtime dependencies are declared for ceph-base (issue#23657, pr#21356, Nathan Cutler)
  • build/ops: rpm,deb: package ceph-kvstore-tool man page (pr#17387, Sage Weil)
  • build/ops: rpm: drop legacy librbd.so.1 symlink in /usr/lib64/qemu (pr#17324, Nathan Cutler)
  • build/ops: rpm: fix _defined_if_python2_absent conditional (pr#20166, Nathan Cutler)
  • build/ops: rpm: fix systemd macros for ceph-volume@.service (issue#22217, pr#19081, Nathan Cutler)
  • build/ops: rpm: move ceph-*-tool binaries out of ceph-test subpackage (issue#21762, pr#18289, Nathan Cutler)
  • build/ops: rpm: Python 3-only ceph-disk and ceph-volume (pr#20140, Nathan Cutler)
  • build/ops: rpm: recommend chrony instead of ntp-daemon (pr#20138, Nathan Cutler)
  • build/ops: rpm: recommend python-influxdb with ceph-mgr (pr#18511, Nathan Cutler, Tim Serong)
  • build/ops: rpm: Revert “ceph.spec: work around build.opensuse.org” (pr#21716, Nathan Cutler)
  • build/ops: rpm: rip out rcceph script (pr#19899, Nathan Cutler)
  • build/ops: rpm: selinux-policy fixes (pr#19026, Brad Hubbard)
  • build/ops: rpm: set build parallelism based on available memory (pr#19122, Nathan Cutler, Richard Brown)
  • build/ops: rpm: set permissions 0755 on rbd resource agent (issue#22362, pr#19494, Nathan Cutler)
  • build/ops: run-make-check.sh: fix SUSE support (issue#22875, pr#20234, Nathan Cutler)
  • build/ops: run-make-check.sh: handle Python 2 absence (issue#23035, pr#20480, Nathan Cutler)
  • build/ops: run-make-check.sh: run ulimit without sudo (pr#17361, yang.wang)
  • build/ops: script/build-integration-branch: print pr url list with titles (pr#17426, Sage Weil)
  • build/ops: selinux: Allow nvme devices (issue#19200, pr#15597, Boris Ranto)
  • build/ops: setup-virtualenv.sh: do not hardcode python binary (issue#23437, pr#21002, Nathan Cutler)
  • build/ops: spdk: update SPDK to fix the build failure on aarch64 (pr#20134, Tone Zhang, Kefu Chai)
  • build/ops: spdk: update SPDK to v17.10 (pr#19208, Kefu Chai)
  • build/ops: spdk: update submodule to more recent upstream (pr#20077, Nathan Cutler)
  • build/ops: specs: require of e2fsprogs (pr#21345, Guillaume Abrioux)
  • build/ops: src/script/build-integration-branch (pr#17382, Sage Weil)
  • build/ops: src: s/pip –use-wheel/pip/ (pr#21159, Kefu Chai)
  • build/ops: submodule: make libmpem as a submodule (pr#17036, Jianpeng Ma)
  • build/ops: sysctl.d: set kernel.pid_max=4194304 (issue#21929, pr#18544, David Disseldorp)
  • build/ops: systemd: rbd-mirror does not start on reboot (pr#17969, Sébastien Han)
  • build/ops: test: delete asok directories correctly (pr#21023, Chang Liu)
  • build/ops: test/fio: enable objectstore FIO plugin building without the need to install and build FIO source code (pr#20535, Igor Fedotov)
  • build/ops,tests: common,test,cmake: various changes to re-enable build on osx (pr#18888, Kefu Chai)
  • build/ops,tests: qa/tests: Changed rhel7.4 to rhel7.5 (pr#21336, Yuri Weinstein)
  • build/ops,tests: test/fio: fix fio objectstore plugin building broken by (pr#20514, Igor Fedotov)
  • build/ops: udev: Fix typo in udev OSD rules file (pr#18976, Mitch Birti)
  • build/ops: use devtoolset-7 on centos/rhel-7 (pr#18863, Kefu Chai)
  • cephfs: Client:Fix readdir bug (pr#18784, dongdong tao)
  • cephfs: Client: setattr should drop “Fs” rather than “As” for mtime and size (pr#18786, dongdong tao)
  • cephfs,common,rbd: common/common_init: disable ms subsystem log gathering for clients (issue#21860, pr#18418, Jason Dillaman)
  • cephfs,common,rbd: Various fixes for SCA issues (pr#21708, Danny Al-Gaaf)
  • cephfs,core: mon/OSDMonitor: set FLAG_SELFMANAGED_SNAPS on cephfs snap removal (issue#23949, pr#21756, Sage Weil)
  • cephfs: MDS: add null check before we push_back “onfinish” (pr#18892, dongdong tao)
  • cephfs: MDS: correct the error msg when init mon client (pr#18836, dongdong tao)
  • cephfs: MDS: make popular counter decay at proper rate (pr#18776, Jianyu Li)
  • cephfs: MDS: make rebalancer evaluate the overload state of each mds with the same criterion (pr#19255, Jianyu Li)
  • cephfs: messages: Initialization of is_primary (pr#16897, amitkuma)
  • cephfs: messages: Initialization of member variables (pr#16898, amitkuma)
  • cephfs: mimic: MDSMonitor: clean up use of pending fsmap in uncommitted ops (issue#23768, pr#22005, Patrick Donnelly)
  • cephfs: mon/MDSMonitor: wait for readable OSDMap before sanitizing (issue#21945, pr#18603, Patrick Donnelly)
  • cephfs,mon: mon/MDSMonitor: fix a bug at preprocess_beacon (pr#17415, wangshuguang)
  • cephfs: osdc/Journaler: use new style options (pr#17806, Kefu Chai)
  • cephfs: qa: check pool full flags (issue#22475, pr#19588, Patrick Donnelly)
  • cephfs: qa: fix typo in test_full (issue#23643, pr#21334, Patrick Donnelly)
  • cephfs: Revert “ceph_context: re-expand admin_socket metavariables in child process” (pr#18545, Patrick Donnelly)
  • cephfs,tests: qa/suites/powercycle/osd/whitelist_health: whitelist slow trimming (pr#17307, Sage Weil)
  • cephfs,tests: qa/workunits/cephtool/test.sh: fix test_mon_mds() (pr#21579, Kefu Chai)
  • cephfs,tools: mount.fuse.ceph: Fix typo (pr#19128, Jos Collin)
  • cephfs: vstart_runner: fixes for recent cephfs changes (pr#19533, Patrick Donnelly)
  • ceph-volume: add ANSIBLE_SSH_RETRIES=5 to functional tests (pr#20592, Andrew Schoen)
  • ceph-volume add functional tests for simple, rearrange lvm tests (pr#18882, Alfredo Deza)
  • ceph-volume: Add linesep/newline at end of JSON file when writing (pr#19458, Wido den Hollander)
  • ceph-volume: adds a –destroy flag to ceph-volume lvm zap (issue#22653, pr#20010, Andrew Schoen)
  • ceph-volume: adds –crush-device-class flag for lvm prepare and create (pr#19949, Andrew Schoen)
  • ceph-volume: adds custom cluster name support to simple (pr#20367, Andrew Schoen)
  • ceph-volume: adds functional CI testing (pr#16919, Andrew Schoen, Alfredo Deza)
  • ceph-volume: adds functional testing for bluestore (pr#18656, Andrew Schoen)
  • ceph-volume: adds raw device support to ‘lvm list’ (issue#23140, pr#20620, Andrew Schoen)
  • ceph-volume: adds success messages for lvm prepare/activate/create (issue#22307, pr#19875, Andrew Schoen)
  • ceph-volume: adds support to zap encrypted devices (issue#22878, pr#20537, Andrew Schoen)
  • ceph-volume: adds the ceph-volume lvm zap subcommand (pr#18513, Andrew Schoen)
  • ceph-volume allow filtering by uuid, do not require osd id (pr#17606, Andrew Schoen, Alfredo Deza)
  • ceph-volume: allow parallel creates (issue#23757, pr#21489, Theofilos Mouratidis)
  • ceph-volume: allow skipping systemd interactions on activate/create (issue#23678, pr#21496, Alfredo Deza)
  • ceph-volume: allow using a device or partition for lvm –data (pr#18924, Alfredo Deza)
  • ceph-volume be resilient to $PATH issues (pr#20650, Alfredo Deza)
  • ceph-volume consume mount/format options from ceph.conf (pr#20408, Alfredo Deza)
  • ceph-volume: correctly fallback to bluestore when no objectstore is specified (pr#19213, Alfredo Deza)
  • ceph-volume correctly normalize mount flags (pr#20543, Alfredo Deza)
  • ceph-volume: create the ceph-volume and ceph-volume-systemd man pages (issue#21030, pr#17152, Alfredo Deza)
  • ceph-volume: dmcrypt support for lvm (issue#22619, pr#20054, Alfredo Deza)
  • ceph-volume dmcrypt support for simple (issue#22620, pr#20264, Andrew Schoen, Alfredo Deza)
  • ceph-volume/doc: add missing subcommand in examples (pr#19381, Guillaume Abrioux)
  • ceph-volume: ensure correct –filestore/–bluestore behavior (pr#18518, Alfredo Deza)
  • ceph-volume failed ceph-osd –mkfs command doesn’t halt the OSD creation process (issue#23874, pr#21685, Alfredo Deza)
  • ceph-volume: fix action plugins path in tests (pr#20910, Guillaume Abrioux)
  • ceph-volume fix filestore OSD creation after mon-config changes (issue#23260, pr#20787, Alfredo Deza)
  • ceph-volume: fix typo in ceph-volume lvm prepare help (pr#21196, Jeffrey Zhang)
  • ceph-volume: fix usage of the –osd-id flag (issue#22642, issue#22836, pr#20203, Andrew Schoen)
  • ceph-volume Format correctly when vg/lv cannot be used (issue#22299, pr#19285, Alfredo Deza)
  • ceph-volume handle inline comments in the ceph.conf file (issue#22297, pr#19319, Alfredo Deza)
  • ceph-volume: handle leading whitespace/tabs in ceph.conf (issue#22280, pr#19259, Alfredo Deza)
  • ceph-volume Implement an ‘activate all’ to help with dense servers or migrating OSDs (pr#21130, Alfredo Deza)
  • ceph-volume improve robustness when reloading vms in tests (pr#21070, Alfredo Deza)
  • ceph-volume: log the current running command for easier debugging (issue#23004, pr#20594, Andrew Schoen)
  • ceph-volume lvm api refactor/move (pr#18110, Alfredo Deza)
  • ceph-volume lvm list (pr#18095, Alfredo Deza)
  • ceph-volume lvm.prepare update to use create_osd_path (pr#18514, Alfredo Deza)
  • ceph-volume: lvm zap will unmount osd paths used by zapped devices (issue#22876, pr#20265, Andrew Schoen)
  • ceph-volume: Nits noticed while studying code (pr#21455, Dan Mick)
  • ceph-volume Persist non-lv devices for journals (pr#17403, Alfredo Deza)
  • ceph-volume process the abspath of the executable first (issue#23259, pr#20824, Alfredo Deza)
  • ceph-volume: removed the explicit use of sudo (issue#22282, pr#19363, Andrew Schoen)
  • ceph-volume: remove extra space (pr#21140, Sébastien Han)
  • ceph-volume rollback on failed OSD prepare/create (issue#22281, pr#19351, Alfredo Deza)
  • ceph-volume should be able to handle multiple LVM (VG/LV) tags (issue#22305, pr#19321, Alfredo Deza)
  • ceph-volume: support GPT and other deployed OSDs (pr#18823, Alfredo Deza)
  • ceph-volume tests add optional flags for vagrant (pr#20849, Alfredo Deza)
  • ceph-volume tests alleviate libvirt timeouts when reloading (issue#23163, pr#20718, Alfredo Deza)
  • ceph-volume tests.devices.lvm prepare isn’t bluestore specific anymore (pr#18984, Alfredo Deza)
  • ceph-volume tests remove unused import (pr#20459, Alfredo Deza)
  • ceph-volume tests use granular env vars for vagrant (pr#20864, Alfredo Deza)
  • ceph-volume: Try to cast OSD metadata to int while scanning directory (pr#19477, Wido den Hollander)
  • ceph-volume update man page for prepare/activate flags (pr#21570, Alfredo Deza)
  • ceph-volume use realpath when checking mounts (issue#22988, pr#20427, Alfredo Deza)
  • ceph-volume: use unique logical volumes (pr#17207, Alfredo Deza)
  • ceph-volume: Using –readonly for {vg|pv|lv}s commands (pr#21409, Erwan Velu)
  • ceph-volume: warn on missing ceph.conf file (issue#22326, pr#19347, Alfredo Deza)
  • ceph-volume warn on mix of filestore and bluestore flags (issue#23003, pr#20513, Alfredo Deza)
  • cleanup: Replacing MIN,MAX with std::min,std::max (pr#18124, Amit Kumar)
  • cli: rados: support for high precision time using stat2 (issue#21199, pr#17395, Abhishek Lekshmanan)
  • cls_acl/_crypto: Add modeline (pr#19010, Shinobu Kinjo)
  • cmake: add chrono to BOOST_COMPONENTS (issue#23424, pr#20977, Nathan Cutler)
  • cmake: add cython_rbd as a dependency to vstart target (pr#18382, Ali Maredia)
  • cmake: bail out if GCC version is less than 5.1 (pr#19344, Kefu Chai)
  • cmake: BuildBoost.cmake: use specified compiler for building boost (pr#19898, Kefu Chai)
  • cmake: bump target jdk to 1.7 (issue#23458, pr#21082, Shengjing Zhu)
  • cmake: bump up required cmake version to 2.8.12 (pr#18285, Kefu Chai)
  • cmake: changes of BuildBoost.cmake to ready seastar (pr#21404, Kefu Chai)
  • cmake: check for aligned_alloc() instead of checking tcmalloc version (pr#18557, Kefu Chai)
  • cmake: check gcc version not release date for libstdc++ saneness (pr#18938, Kefu Chai)
  • cmake: check version of boost in src/boost (pr#19914, Kefu Chai)
  • cmake: cleanups (pr#18597, Kefu Chai)
  • cmake,common: changes to port part of ceph to osx (pr#17615, Kefu Chai)
  • cmake: compile nvml as an external project (pr#17462, Jianpeng Ma)
  • cmake: define HAVE_STDLIB_MAP_SPLICING for both libstdc++ and libc++ (pr#21284, Kefu Chai)
  • cmake: disable DOWNLOAD_NO_PROGRESS if cmake ver is lower than 3.1 (pr#20492, Kefu Chai)
  • cmake: disable FAIL_ON_WARNINGS for rocksdb (pr#19426, Kefu Chai)
  • cmake: disable VTA on options.cc (pr#17393, Kefu Chai)
  • cmake: do not find bzip2/lz4 for rocksdb (pr#19963, runsisi)
  • cmake: do not link against librados.a (pr#18576, Kefu Chai)
  • cmake: do not link against unused or duplicated libraries (pr#18092, Kefu Chai)
  • cmake: enabled py3 only build (pr#20064, Kefu Chai)
  • cmake: enable LZ4 by default (pr#21332, Grant Slater, Casey Bodley)
  • cmake: enable new policies to silence cmake warnings (pr#21662, Kefu Chai)
  • cmake: fix building without mgr module (pr#21591, Yuan Zhou)
  • cmake: fix frontend cmake build (pr#21449, Ricardo Dias)
  • cmake: fix libcephfs-test.jar build failure (issue#22828, pr#20175, Tone Zhang)
  • cmake: fix the include dir for building boost::python (pr#20324, Kefu Chai)
  • cmake: fix typo in status message (pr#21464, Lenz Grimmer)
  • cmake: hide symbols import from other libraries in libcls_* (issue#23517, pr#21571, Kefu Chai)
  • cmake: identify the possible incompatibility of rocksdb and tcmalloc (issue#21422, pr#17788, Kefu Chai)
  • cmake: in case of bad “ALLOCATOR” selected issue warning (pr#17422, Adam Kupczyk)
  • cmake: include frontend build in ‘all’ target (pr#21466, John Spray)
  • cmake: let “tests” depend on “mgr-dashboard-frontend-build” (pr#21468, Kefu Chai)
  • cmake: ‘make check’ builds radosgw and its cls dependencies (pr#20422, Casey Bodley)
  • cmake: mgr: exclude .gitignore (pr#19174, Nathan Cutler)
  • cmake/modules/BuildRocksDB.cmake: enable compressions for rocksdb (issue#24025, pr#22183, Kefu Chai)
  • cmake: only create sysctl file on linux (pr#19029, Kefu Chai)
  • cmake: pass static linkflags to the linker who links libcommon (pr#19763, Kefu Chai)
  • cmake: s/boost_256/boost_sha256/ (pr#21573, Kefu Chai)
  • cmake: set supported language the right way (pr#18216, Kefu Chai)
  • cmake: should use the value of GPERFTOOLS_LIBRARIES as REQUIRED_VARS (pr#18645, Kefu Chai)
  • cmake: s/sysconf/sysconfig/ (pr#20631, Kefu Chai)
  • cmake: sync nvml submodule to latest code (pr#20411, Jianpeng Ma)
  • cmake: System Includes to silence warnings from submodules and libraries! (pr#18711, Adam C. Emerson)
  • cmake: typo fix when npm is not found (pr#20801, Abhishek Lekshmanan)
  • cmake: update minimum boost version to 1.66 (issue#20048, issue#22600, pr#19808, Casey Bodley)
  • cmake: update the error message for gperftools bug (pr#17901, Kefu Chai)
  • cmake: warn if libstdc++ older than 5.1.0 is used (pr#18837, Kefu Chai)
  • cmake: WITH_SPDK=ON by default (pr#18944, Liu-Chunmei, Kefu Chai, wanjun.lp, Ziye Yang)
  • common: adding line break at end of some cli results (issue#21019, pr#16687, songweibin)
  • common: add line break for “ceph daemon TYPE.ID version” (pr#17146, Zhu Shangzhong)
  • common: Add metadata with only Ceph version number and release (pr#21095, Wido den Hollander)
  • common: Add min/max of ms_async_op_threads (pr#19942, Shinobu Kinjo)
  • common: Add noreturn attribute to silence uninitialized warning (pr#19348, Adam C. Emerson)
  • common: auth: add err reason for log info in load function (pr#17256, Luo Kexue)
  • common: bench test fall into dead loop when <seconds>=0 (pr#16382, PC)
  • common: buffer: avoid changing bufferlist ABI by removing new _mempool field (issue#21573, pr#18408, Sage Weil)
  • common: by default, do not assert on leaks in the shared_cache code (issue#21737, pr#18201, Greg Farnum)
  • common: ceph: add the right bracket to watch-channel argument in the help message (pr#19698, Chang Liu)
  • common: ceph.in: execv using the same python (pr#17713, Kefu Chai)
  • common: ceph_release: s/rc/stable/ (pr#22264, Sage Weil)
  • common: change routines to public access (pr#20003, Willem Jan Withagen)
  • common: Check this->data.op_size before use (pr#18816, Amit Kumar)
  • common: cleanup address_helper (pr#19643, Shinobu Kinjo)
  • common: cmake,common/RWLock: check for libpthread extensions (pr#19202, Kefu Chai)
  • common: common: add for_each_substr() for cheap string split (pr#18798, Casey Bodley)
  • common: common: add streaming interfaces for json/xml escaping (pr#19806, Casey Bodley)
  • common: common/admin_socket: validate command json before feeding it to hook (pr#20437, Kefu Chai)
  • common: common/blkdev: fix build in FreeBSD environment (pr#19316, Mykola Golub)
  • common: common/buffer: cleanups (pr#18312, Shinobu Kinjo)
  • common: common/buffer: switch crc cache to single pair instead of map (pr#18906, Piotr Dałek)
  • common: common/config: add units to options (issue#22747, pr#20419, Kefu Chai)
  • common: common/config: limit calls to normalize_key_name (pr#20318, Piotr Dałek)
  • common: common/config: make internal_safe_to_start_threads internal (pr#18884, Sage Weil)
  • common: common/ConfUtils: check key before actually normalizing (pr#20370, Piotr Dałek)
  • common: common/dns_resolv.cc: Query for AAAA-record if ms_bind_ipv6 is True (issue#23078, pr#20530, Wido den Hollander)
  • common: common/dns_resolve: fix memory leak (pr#19649, Yao Zongyou)
  • common: common/event_socket.h: include <errno.h> to use errno (pr#18351, Kefu Chai)
  • common: common/Formatter: fix string_view usage for {json,xml}_stream_escaper (issue#23622, pr#21317, Sage Weil)
  • common: common/hobject: compare two objects’ key directly (pr#21062, xie xingguo)
  • common: common/hobject: preserve the order of hobject (pr#21217, Kefu Chai)
  • common: common/ipaddr: Do not select link-local IPv6 addresses (issue#21813, pr#20862, Wido den Hollander)
  • common: common/lockdep: drop hash<pthread_t> specialization (pr#20574, Kefu Chai)
  • common: common/LogClient: assign seq and queue atomically (issue#18209, pr#16828, Sage Weil)
  • common: common/log: Speed improvement for log (pr#19100, Adam Kupczyk, Kefu Chai)
  • common: common/OpHistory: move insert/cleanup into separate thread (pr#20540, Piotr Dałek)
  • common: common/options: drop unused options (pr#20895, Kefu Chai)
  • common: common/options: long description for log_stderr_prefix (pr#19869, Sage Weil)
  • common: common/options: pass by reference and use user-literals for size (pr#18034, Kefu Chai)
  • common: common/options: use user-defined literals for default values (pr#17180, Kefu Chai)
  • common: common/perf_counters: remove unused parameter (pr#19805, Kefu Chai)
  • common: common/pick_address.cc: Cleanup (pr#19707, Shinobu Kinjo)
  • common: common/pick_address: wrong prefix_len in pick_iface() (pr#20128, Gu Zhongyan)
  • common: common/str_list: s/boost::string_view/std::string_view (pr#20475, Kefu Chai)
  • common: common/strtol: fix strict_strtoll() so it accepts hex starting with 0x (pr#21521, Kefu Chai)
  • common: common/strtoll: remove superfluous const modifier (pr#21560, Jan Fajerski)
  • common: common/throttle: start using 64-bit values (issue#22539, pr#19759, Igor Fedotov)
  • common: common/types: make numbers a bit nicer when displaying space usage (pr#17126, xie xingguo)
  • common: common/util: do not print error if VERSION_ID is missing (pr#17787, Kefu Chai)
  • common: compressor: use generate_random_number() for type=”random” (pr#18272, Casey Bodley)
  • common: compressor/zstd: improvements (pr#18879, Sage Weil)
  • common: compute SimpleLRU’s size with contents.size() instead of lru.size() (issue#22613, pr#19813, Xuehan Xu)
  • common: config: expand tilde for ~/.ceph/$cluster.conf (issue#23215, pr#20774, Rishabh Dave)
  • common: config: notify config observers on set_mon_vals() (pr#21161, Casey Bodley)
  • common: config: Remove _get_val (pr#18222, Adam C. Emerson)
  • common/config: use with_val() for better performance (pr#19056, Adam C. Emerson)
  • common: consolidate spinlocks (pr#15816, Jesse Williamson)
  • common,core: common, osd: various cleanups (pr#18149, Kefu Chai)
  • common,core: common/pick_address: add {public,cluster}_network_interface option (pr#18028, Sage Weil)
  • common,core: common/Throttle: Clean up (pr#16618, Adam C. Emerson)
  • common,core: fix broken use of streamstream::rdbuf() (issue#22715, pr#19998, Sage Weil)
  • common,core: include/ceph_features: deprecate a bunch of features (pr#18546, Sage Weil)
  • common,core: include,messages,rbd: Initialize counter,group_pool (pr#17774, Amit Kumar)
  • common,core: options: Do not use linked lists of pointers! (pr#17984, Adam C. Emerson)
  • common,core: osdc/Objecter: take budgets across a LingerOp instead of on child Ops (issue#22882, pr#20519, Greg Farnum)
  • common,core: osd/OpRequest: reduce overhead when disabling tracking (pr#18470, Haomai Wang)
  • common,core: rados: Prefer templates to macros (pr#19913, Adam C. Emerson)
  • common,core,rbd,rgw: common,osd,rgw: Fixes for issues found during SCA (pr#21419, Danny Al-Gaaf)
  • common,core,rbd,tests,tools: common,mds,mgr,mon,osd: store event only if it’s added (pr#16312, Kefu Chai)
  • common,core: Revert “msg/async/AsyncConnection: unregister connection when racing happened” (issue#22231, pr#19586, Sage Weil)
  • common,core: Revert “osd/OSDMap: allow bidirectional swap of pg-upmap-items” (issue#21410, pr#17760, Sage Weil)
  • common: Coverity and SCA fixes (pr#17431, Danny Al-Gaaf)
  • common/crc/aarch64: Added cpu feature pmull and make aarch64 specific… (pr#22184, Adam Kupczyk)
  • common: crush/CrushWrapper: fix out of bounds access (issue#20926, pr#16869, Sage Weil)
  • common: crypto: remove cryptopp library (pr#20015, Casey Bodley)
  • common: denc cleanups and other fixes (pr#19877, Adam C. Emerson)
  • common: denc: support enum with underlying type (pr#18701, Kefu Chai)
  • common: Destroy attr of RWLock after initialized (pr#17103, Wen Zhang)
  • common: dmclock: update mClockPriorityQueue with changes in subtree (pr#20992, Casey Bodley)
  • common: dout: DoutPrefixProvider operates directly on stream (pr#21608, Casey Bodley)
  • common: drop namespace using directives for std (pr#19159, Shinobu Kinjo)
  • common: drop unused variables “bluestore_csum_*_block” in opts (pr#17394, songweibin)
  • common: encoding: reset optional<> if it is uninitialized (pr#17599, Kefu Chai)
  • common: Extends random.h: numeric types relaxed to compatible types (with (pr#20670, Jesse Williamson)
  • common: fix BoundedKeyCounter const_pointer_iterator (issue#22139, pr#18953, Casey Bodley)
  • common: fix daemon abnormal exit at parsing invalid arguments (issue#21365, issue#21338, pr#17664, Yan Jun)
  • common: fix potential memory leak in HTMLFormatter (pr#20699, Yao Zongyou)
  • common: fix typo deamon in comments (pr#17687, yonghengdexin735)
  • common: fix typo in options.cc (pr#20549, songweibin)
  • common: FreeBSD wants the correct struct selection for ipv6 (issue#21813, pr#21143, Willem Jan Withagen)
  • common: global: output usage on -h, –help, or no args before contacting mons (pr#20812, Sage Weil)
  • common: hint the main branch of dout() accordingly to default verbosity (pr#21259, Radoslaw Zarzynski)
  • common: implement random number generator (following N3551) (issue#18873, pr#15341, Jesse Williamson)
  • common: Improving message sent to user when getting signals (issue#23320, pr#21000, Erwan Velu)
  • common: include/encoding: fix compat version error message (pr#19660, Xinying Song)
  • common: include/interval_set: parameterize by map type and kill btree_interval_set.h (pr#18611, Sage Weil)
  • common: include/rados: fix typo in librados.h (pr#17988, wumingqiao)
  • common: include: Remove unused header, ciso646 (pr#18320, Shinobu Kinjo)
  • common: include/types: format decimal numbers with decimal factor (issue#22095, pr#19117, Jan Fajerski)
  • common: include: xlist: Fix Clang error for missing string (pr#19367, Willem Jan Withagen)
  • common: interval_set: kill subset_of() (pr#21108, xie xingguo)
  • common: interval_set: optimize intersect_of insert operations (issue#21229, pr#17265, Zac Medico)
  • common: introduce md_config_cacher_t (pr#20320, Radoslaw Zarzynski)
  • common: kick off mimic (pr#16993, Sage Weil)
  • common: lockdep fixes (issue#20988, pr#17738, Jeff Layton)
  • common: log: clear thread-local stream’s ios flags on reuse (pr#20174, Casey Bodley)
  • common: logically dead code inside shunique_lock.h (pr#17341, Amit Kumar)
  • common: make ceph_clock_now() inlineable (pr#20443, Radoslaw Zarzynski)
  • common: Make code to invoke assert() smaller (pr#20445, Adam Kupczyk)
  • common: make some message informative, instead of error (pr#16594, Willem Jan Withagen)
  • common: mark events of TrackedOp outside its constructor (issue#22608, pr#19828, Xuehan Xu)
  • common: mgr/dashboard_v2: Fix test_cluster_configuration test (issue#23265, pr#20782, Sebastian Wagner)
  • common: mimic: include/types: space between number and units (pr#22107, Sage Weil)
  • common,mon: crush,mon: fix weight-set vs crush device classes (issue#20939, pr#16883, Sage Weil)
  • common,mon,osd,pybind: silence warning and remove executable mode bit (pr#17512, Kefu Chai)
  • common: msg/async/AsyncConnection: less noisy debug (pr#20600, Sage Weil)
  • common: msg/async: execute on core specified by core_id not its index (pr#20659, Kefu Chai)
  • common: msg/msg_types: fix the entity_addr_t’s decoder (pr#17699, Kefu Chai)
  • common: msg/simple: s/ceph::size/std::size/ (pr#19896, Kefu Chai)
  • common/options.cc: cleanup readable literals for default sizes (pr#18425, Enming Zhang)
  • common/options.cc: Set Filestore rocksdb compaction readahead option (issue#21505, pr#17900, Mark Nelson)
  • common: OpTracker doesn’t visit TrackedOp when nref == 0 (issue#24037, pr#22160, Radoslaw Zarzynski)
  • common: osdc/Objecter: fix warning (pr#21757, Sage Weil)
  • common: osdc/Objecter: record correctly value for l_osdc_op_send_bytes (issue#21982, pr#18810, Jianpeng Ma)
  • common: osd/PrimaryLogPG: send requests to primary on cache miss (issue#20919, pr#16884, Sage Weil)
  • common: osd_types: define max in eversion_t::max() to static (pr#17453, yang.wang)
  • common,os: initialize commit_data,cmount,iocb (pr#17766, Amit Kumar)
  • common: posix_fallocate on ZFS returns EINVAL (pr#20398, Willem Jan Withagen)
  • common: rados: clean up rados_getxattrs() and rados_striper_getxattrs() (pr#20259, Gu Zhongyan)
  • common: RAII-styled mechanism for updating PerfCounters (pr#19149, Radoslaw Zarzynski)
  • common: random: revert change from boost::optional to std::optional (issue#23778, pr#21567, Casey Bodley)
  • common: Remove ceph_clock_gettime, extern keyword (pr#19353, Shinobu Kinjo)
  • common: retry_sys_call no need take address of a function pointer (pr#21281, Leo Zhang)
  • common: Revert “common/config: return const reference instead of a copy” (pr#18934, Kefu Chai)
  • common: Revert “core: hint the dout()’s message crafting as a cold code.” (issue#23169, pr#20636, Kefu Chai)
  • common,rgw: rgw,common,rbd: s/boost::regex/std::regex/ (pr#19393, Kefu Chai)
  • common,rgw: rgw,common: remove already included header files (pr#19390, Yao Zongyou)
  • common: silence jenkins’s buiding warning in obj_bencher.cc (pr#17272, Luo Kexue)
  • common: src/common: update some ms_* options to be more consistent (pr#20652, shangfufei)
  • common: src/msg/async/rdma: decrease cpu usage by rdtsc instruction (pr#16965, Jin Cai)
  • common: Static Pointer (pr#19079, Adam C. Emerson)
  • common: strict_strtol INT_MAX and INT_MIN is valid (pr#18574, Shasha Lu)
  • common: s/unique_lock/lock_guard/, if manual lock/unlock are not necessary (pr#19770, Shinobu Kinjo)
  • common: Switch singletons to use immobile_any and cleanups (pr#20273, Adam C. Emerson)
  • common: test: fix unittest memory leak to silence valgrind (pr#19654, Yao Zongyou)
  • common,tests: test/common: unittest_mclock_priority_queue builds with “make” command (pr#17582, J. Eric Ivancich)
  • common,tests: test/librados: create unique lock names (issue#20798, pr#16953, Neha Ojha)
  • common: tools/crushtool: skip device id if no name exists (issue#22117, pr#18901, Jan Fajerski)
  • common: use mono clock for HeartbeatMap (pr#17827, Xinze Chi, Kefu Chai)
  • common: use move instead of copy in build_options() (pr#18003, Casey Bodley)
  • common: utime: fix __32u sec time overflow (pr#21113, kungf)
  • compressor: add zstd back (pr#21106, Kefu Chai)
  • compressor: conditionalize on HAVE_LZ4 (pr#17059, Kefu Chai)
  • compressor: kill AsyncCompressor which is broken (pr#18472, Haomai Wang)
  • core: blkin: Fix unconditional tracing in OSD (pr#19156, Yingxin)
  • core: ceph-debug-docker.sh: add ceph-osd-dbg package (pr#17947, Patrick Donnelly)
  • core: ceph.in: Add blocking mode for scrub and deep-scrub (pr#19793, Brad Hubbard)
  • core: ceph.in: do not panic at control+d in interactive mode (pr#18374, Kefu Chai)
  • core: ceph.in: print all matched commands if arg missing (issue#22344, pr#19547, Kefu Chai)
  • core: ceph.in: use a different variable for holding thrown exception (pr#20663, Kefu Chai)
  • core: ceph-kvstore-tool: copy to different store type and cleanup properly (pr#18029, Josh Durgin)
  • core: ceph-mgr: exit after usage (issue#23482, pr#21401, Sage Weil)
  • core: ceph_osd.cc: Drop legacy or redundant code (pr#18718, Shinobu Kinjo)
  • core: ceph-osd: some flags are not documented in the help output (issue#20057, pr#15565, Yanhu Cao)
  • core: ceph: print output of “status” as string not as bytes (pr#21297, Kefu Chai)
  • core: ceph-rest-api: when port=0 use the DEFAULT_PORT instead (pr#17443, You Ji)
  • core: ceph_test_objectstore: disable filestore_fiemap for tests (issue#21880, pr#18452, Sage Weil)
  • core: ceph_test_objectstore: do not change model for 0-length zero (issue#21712, pr#18519, Sage Weil)
  • core: ceph_test_rados_api_aio: fix race with full pool and osdmap (issue#23916, issue#23917, pr#21709, Sage Weil)
  • core: ceph_test_rados_api_tier: add ListSnap test (pr#17706, Xuehan Xu)
  • core: client,osd,test: Initialize fuse_req_key,snap,who,seq (pr#17772, Amit Kumar)
  • core: common/admin_socket: various cleanups (pr#20274, Adam C. Emerson)
  • core: common/config: cleanup remove some unused macros (pr#19599, Yao Zongyou)
  • core: common,mds,osd: Explicitly delete copy ctor if noncopyable (pr#19465, Shinobu Kinjo)
  • core: common/options: enable multiple rocksdb compaction threads for filestore (pr#18232, Josh Durgin)
  • core: common, osd: duplicated “start” event in OpTracker, improve OpTracker::dump_ops (pr#21119, Chang Liu)
  • core: compressor: Add Brotli Compressor (pr#19549, BI SHUN KE)
  • core: config: lower default omap entries recovered at once (issue#21897, pr#19910, Josh Durgin)
  • core: crush/CrushWrapper: fix potential invalid use of iterator (pr#21325, xie xingguo)
  • core: dmclock: Delivery of the dmclock delta, rho and phase parameter + Enabling the client service tracker (pr#16369, Byungsu Park, Taewoong Kim)
  • core: erasure-code: refactor the interfaces to hide internals from public (pr#18683, Kefu Chai)
  • core: erasure-code: use jerasure_free_schedule to properly free a schedule (pr#19650, Yao Zongyou)
  • core: erasure-code: use std::count() instead (pr#19428, Kefu Chai)
  • core: etc/default/ceph: remove jemalloc option (issue#20557, pr#18486, Sage Weil)
  • core: filestore: include <linux/falloc.h> (pr#20415, wumingqiao)
  • core: Fix a dead lock when doing rdma performance test by fio (pr#17016, Wang Chuanhong)
  • core: Fix asserts caused by DNE pgs left behind after lots of OSD restarts (issue#21833, pr#20571, David Zafman)
  • core: include: kill MIN and MAX macros (pr#20886, Sage Weil)
  • core: interval_set: optimize intersection_of (pr#17088, Zac Medico)
  • core: kv/KeyValueDB: add column family (pr#18049, Jianjian Huo, Adam C. Emerson, Sage Weil)
  • core: kv/RocksDB: get index and filter blocks memory usage (pr#19934, Zhi Zhang)
  • core: kv/RocksDBStore: fix rocksdb error when block cache is disabled (issue#23816, pr#21583, Yang Honggang)
  • core: librados: add OPERATION_ORDERSNAP flag and yet another aio_operate method (pr#20343, Mykola Golub)
  • core: librados.h: add LIBRADOS_SUPPORTS_APP_METADATA (pr#16542, Matt Benjamin)
  • core: libradosstriper: fix the function declaration of rados_striper_trunc (pr#20301, yuelongguang)
  • core: libradosstriper: silence warning from -Wreorder (pr#16890, songweibin)
  • core: make the main dout() paths faster and more maintanable (pr#20290, Radoslaw Zarzynski)
  • core: messages: Initialization of variable beat (pr#17641, Amit Kumar)
  • core: messages: Initialize member variables (pr#16846, amitkuma)
  • core: messages: initialize variable tid in MMDSFindIno (pr#16793, amitkuma)
  • core: messages: Initializing members in MOSDPGUpdateLogMissing (pr#16928, amitkuma)
  • core: messages: Initializing variable ceph_mds_reply_head (pr#17090, amitkuma)
  • core: messages,journal: Initialization of stats_period,m_active_set (pr#17792, Amit Kumar)
  • core: messages/MOSDMap: do compat reencode of crush map, too (issue#21882, pr#18454, Sage Weil)
  • core: messages/MOSDOp: a fixes of encode_payload (pr#16836, Ying He)
  • core: messages: Silence uninitialized member warnings (pr#17596, Amit Kumar)
  • core: mgr/DaemonServer.cc: add ‘is_valid=false’ when decode caps error (issue#20990, pr#16978, Yanhu Cao)
  • core,mgr: mgr/balancer: improve error message (issue#22814, pr#21427, Sage Weil)
  • core,mgr: osd,mgrclient: pass daemon_status by rvalue ref and other cleanups (pr#18509, Kefu Chai)
  • core,mgr: osd,mgr: report slow requests and pending creating pgs to mgr (pr#18614, Kefu Chai)
  • core: mimic: crush: update choose_args on bucket removal (issue#24167, pr#22120, Sage Weil)
  • core: mimic: osdc: guard op->on_notify_finish with lock (issue#23966, pr#21834, Kefu Chai)
  • core: mimic: osd: clean up smart probe (issue#23899, issue#24104, pr#21959, Sage Weil, Gu Zhongyan)
  • core: mimic: osd: Don’t evict even when preemption has restarted with smaller chunk (pr#22041, David Zafman)
  • core: mimic: osd/PrimaryLogPG: fix try_flush_mark_clean write contention case (issue#24200, issue#24174, pr#22113, Sage Weil)
  • core: mon/ConfigKeyService: dump: print placeholder value for binary blobs (issue#23622, pr#21329, Sage Weil)
  • core,mon: crush, mon: bump up map version only if we truly created a weight-set (pr#20178, xie xingguo)
  • core: mon/LogMonitor: separate out summary by channel (pr#21395, Sage Weil)
  • core,mon: mon/AuthMonitor: create bootstrap keys on create_initial() (pr#21236, Joao Eduardo Luis)
  • core,mon: mon/LogMonitor: do not crash on log sub w/ no messages (pr#21469, Sage Weil)
  • core,mon: mon,osd,crush: misc cleanup (pr#20687, songweibin)
  • core,mon: mon/OSDMonitor: Comment out unused function (pr#20275, Brad Hubbard)
  • core,mon: mon/OSDMonitor: don’t create pgs if pool was deleted (issue#21309, pr#17600, Joao Eduardo Luis)
  • core,mon: mon/OSDMonitor: implement cluster pg limit (pr#17427, Sage Weil)
  • core,mon: mon/OSDMonitor: list osd tree in named bucket (pr#19564, kungf)
  • core: mon, osd: add create-time for pool (pr#21690, xie xingguo)
  • core: mon, osd: fix potential collided *Up Set* after PG remapping (issue#23118, pr#20653, xie xingguo)
  • core,mon: osd,mon: add max-pg-per-osd limit (pr#18358, Kefu Chai)
  • core: mon/OSDMonitor: filter out pgs that shouldn’t exist from force-create-pg (pr#20267, Sage Weil)
  • core: mon/OSDMonitor: fix min_size default for replicated pools (pr#20555, Josh Durgin)
  • core: mon/OSDMonitor: Fix OSDMonitor error message outputs (issue#22351, pr#20022, Brad Hubbard)
  • core: mon/OSDMonitor: make ‘osd crush class rename’ idempotent (pr#17330, xie xingguo)
  • core: mon/OSDMonitor: rename outer name declaration to avoid shadowing (pr#20032, Sage Weil)
  • core: mon/OSDMonitor: require –yes-i-really-mean-it for force-create-pg (pr#21619, Sage Weil)
  • core: mon,osd,osdc: refactor snap trimming (phase 1) (pr#18276, Sage Weil)
  • core: mon, osd: per pool space-full flag support (pr#17371, xie xingguo)
  • core: mon, osd: turn down non-error scrub message severity (issue#20947, pr#16916, John Spray)
  • core: mon/PGMap: fix PGMapDigest decode (pr#22099, Sage Weil)
  • core: mon/PGMap: Fix %USED calculation bug (issue#22247, pr#19165, Xiaoxi Chen)
  • core: mon/PGMap: remove or narrow columns ‘pg ls’ output (pr#20945, Sage Weil)
  • core: mon/PGMap: ‘unclean’ does not imply damaged (pr#18493, Sage Weil)
  • core: MOSDPGRecoveryDelete[Reply]: bump header version (pr#17585, Josh Durgin)
  • core: msg/asyc/rmda: fix the bug of assert when Infiniband::recv_msg receives disconnect message (pr#17688, Jin Cai)
  • core: msg/async/AsyncConnection: combine multi alloc into one (pr#18833, Haomai Wang)
  • core: msg/async/AsyncConnection: Fix FPE in process_connection (issue#23618, pr#21314, Brad Hubbard)
  • core: msg/async/AsyncConnection: state will be NONE if replacing by another one (issue#21883, pr#18467, Haomai Wang)
  • core: msg/async/AsyncConnection: unregister connection when racing happened (pr#19013, Haomai Wang)
  • core: msg/async: batch handle numevents (pr#18321, Jianpeng Ma)
  • core: msg/async: don’t kill connection if replacing (issue#21143, pr#17288, Haomai Wang)
  • core: msg/async: don’t stuck into resetsession/retrysession loop (pr#17276, Haomai Wang)
  • core: msg/async: fix bug of data type conversion when uint64_t -> int -> uint64_t (pr#18210, shangfufei)
  • core: msg/async: print error log if add_event fail (pr#17102, mychoxin)
  • core: msg/async/rdma: fix multi cephcontext confllicting (pr#16893, Haomai Wang)
  • core: msg/async/rdma: fix the bug that rdma polling thread uses the same thread name with msg worker (pr#16936, Jin Cai)
  • core: msg/async/rdma: improves RX buffer management (pr#16693, Alex Mikheev)
  • core: msg/async/rdma: uninitialized variable fix (pr#18091, Vasily Philipov)
  • core: msg/DispatchQueue: clear queue after wait() (issue#18351, pr#20374, Sage Weil)
  • core: msgr/simple: set Pipe::out_seq to in_seq of the connecting side (issue#23807, pr#21585, Xuehan Xu)
  • core: os/bluestore: debug bluestore cache shutdown (issue#21259, pr#17844, Sage Weil)
  • core: os/bluestore: disable on_applied sync_complete (issue#22668, pr#20169, Sage Weil)
  • core: os/bluestore: make bdev label parsing error more meaningful and less noisy (pr#20090, Sage Weil)
  • core: os/bluestore: make BlueStore opened by start_kv_only umountable (issue#21624, pr#18082, Chang Liu)
  • core: os/bluestore: use db->rm_range_keys to delete range of keys (pr#18279, Xiaoyan Li)
  • core: OSD/admin_socket: add get_mapped_pools command (pr#19112, Xiaoxi Chen)
  • core: osdc, class_api: kill implicit string conversions (pr#16648, Piotr Dałek)
  • core: osdc: dec num_in_flight for pool_dne case (pr#21110, Jianpeng Ma)
  • core: osdc: Do not use lock_guard as unique_lock (pr#19756, Shinobu Kinjo)
  • core: osdc: invoke notify finish context on linger commit failure (issue#23966, pr#21786, Jason Dillaman)
  • core: osdc/Objecter: add ignore overlay flag if got redirect reply (pr#21275, Ting Yi Lin)
  • core: osdc/Objecter: delay initialization of hobject_t in _send_op (issue#21845, pr#18427, Jason Dillaman)
  • core: osdc/Objecter: fix recursive locking in _finish_command (issue#23940, pr#21742, Sage Weil)
  • core: osdc/Objecter: misc cleanups (pr#18476, Jianpeng Ma)
  • core: osdc/Objecter: prevent double-invocation of linger op callback (issue#23872, pr#21649, Jason Dillaman)
  • core: osdc/Objecter: skip sparse-read result decode if bufferlist is empty (issue#21844, pr#18400, Jason Dillaman)
  • core: osd,compressor: Expose compression algorithms via MOSDBoot (issue#22420, pr#20558, Jesse Williamson)
  • core: osdc: remove unused function (pr#21081, Jianpeng Ma)
  • core: osd,dmclock: use pointer to ClientInfo instead of a copy of it (pr#18387, Kefu Chai)
  • core: osd: do not forget pg_stat acks which failed to send (pr#16702, huangjun)
  • core: OSD: drop unsed parameter passed to check_osdmap_features (pr#18466, Leo Zhang)
  • core: osd/ECBackend: inject sleep during deep scrub (pr#20531, xie xingguo)
  • core: osd/ECBackend: only check required shards when finishing recovery reads (issue#23195, pr#21273, Josh Durgin)
  • core: osd/ECBackend: update misleading comment about EIO handling (pr#21686, Josh Durgin)
  • core: osd/ECBackend: wait for apply for luminous peers (pr#21604, Sage Weil)
  • core: osd/ECMsgTypes: fix ECSubRead compat decode (pr#20948, Sage Weil)
  • core: osd, librados: add a rados op (TIER_PROMOTE) (pr#19362, Myoungwon Oh)
  • core: osd,librados: add manifest, operations for chunked object (pr#15482, Myoungwon Oh)
  • core: osd,messages: Initialize read_length,options,send_reply (pr#17799, Amit Kumar)
  • core: osd/OSD: batch-list objects to reduce memory consumption (pr#20767, xie xingguo)
  • core: osd/OSD.cc: add ‘isvalid=false’ when failed to parse caps (pr#16888, Yanhu Cao)
  • core: osd/OSD.cc: use option ‘osd_scrub_cost’ instead (pr#18479, Liao Weizhong)
  • core: osd/OSDMap: add osdmap epoch info when printing info summary (pr#20184, shun-s)
  • core: osd/OSDMap: fix HAVE_FEATURE logic in encode() (pr#20922, Ilya Dryomov)
  • core: osd/OSDMap: ignore PGs from pools of failure-domain OSD (pr#20703, xie xingguo)
  • core: osd/OSDMap: misleading message in print_oneline_summary() (issue#22350, pr#20313, Gu Zhongyan)
  • core: osd/OSDMap: more pg upmap fixes (issue#23878, pr#21670, xiexingguo)
  • core: osd/OSDMap: remove the unnecessary checks for null (pr#18636, Kefu Chai)
  • core: osd/OSDMap: skip out/crush-out osds (pr#20655, xie xingguo)
  • core: osd/OSDMap: upmap should respect the osd reweights (issue#21538, pr#17944, Theofilos Mouratidis)
  • core: osd/osd_type: get_clone_bytes - inline size() for overlapping size (pr#17823, xie xingguo)
  • core: osd/osd_types.cc: copy extents map too while making clone (pr#18396, xie xingguo)
  • core: osd/osd_types: fix ideal lower bound object-id of pg (pr#21235, xie xingguo)
  • core: osd/osd_types: fix object_stat_sum_t decode (pr#18551, Sage Weil)
  • core: osd/osd_types: fix pg_pool_t encoding for hammer (pr#21282, Sage Weil)
  • core: osd/osd_types: kill preferred field in pg_t (pr#20567, Sage Weil)
  • core: osd/osd_types: object_info_t: remove unused function (pr#17905, Kefu Chai)
  • core: osd/osd_types: pg_pool_t: remove crash_replay_interval member (pr#18379, Sage Weil)
  • core: osd/osd_types: remove backlog type for pg_log_entry_t (pr#20887, Sage Weil)
  • core: osd/OSD: Using Wait rather than WaitInterval to wait queue.is_empty() (pr#17659, Jianpeng Ma)
  • core: osd/PG: allow scrub preemption (pr#18971, Sage Weil)
  • core: osd/PGBackend: delete reply if fails to complete delete request (issue#20913, pr#17183, Kefu Chai)
  • core: osd/PGBackend: drop input “snapid_t” from objects_list_range() (pr#21151, xie xingguo)
  • core: osd/PGBackend: fix large_omap_objects checking (pr#21150, xie xingguo)
  • core: osd/PGBackend: release a msg using msg->put() not delete (issue#20913, pr#17246, Kefu Chai)
  • core: osd/PG: const cleanup for recoverable/readable predicates (pr#18982, Neha Ojha)
  • core: osd/PG: decay scrub_chunk_max too if scrub is preempted (pr#20552, xie xingguo)
  • core: osd/PG: discard msgs from down peers (issue#19605, pr#17217, Kefu Chai)
  • core: osd/PG: drop unused variable “oldest_update” in PG.h (pr#17142, songweibin)
  • core: osd/PG: extend pg state bits to fix pg ls commands error (issue#21609, pr#18058, Yan Jun)
  • core: osd/PG: fix calc of misplaced objects (pr#18528, Kefu Chai)
  • core: osd/PG: fix DeferRecovery vs AllReplicasRecovered race (issue#23860, pr#21706, Sage Weil)
  • core: osd/PG: fix objects degraded higher than 100% (issue#21803, issue#21898, pr#18297, Sage Weil, David Zafman)
  • core: osd/PG: fix out of order priority for PG deletion (pr#21613, xie xingguo)
  • core: osd/PG: fix recovery op leak (pr#18524, Sage Weil)
  • core: osd/PG: fix uninit read in Incomplete::react(AdvMap&) (issue#23980, pr#21798, Sage Weil)
  • core: osd/PG: force rebuild of missing set on jewel upgrade (issue#20958, pr#16950, Sage Weil)
  • core: osd/PG: include primary in PG operator<< for ec pools (pr#19453, Sage Weil)
  • core: osd/PGLog: assert out on performing overflowed log trimming (pr#21580, xie xingguo)
  • core: osd/PGLog: cleanup unused function revise_have (pr#19329, Enming Zhang)
  • core: osd/PGLog: fix sanity check against **complete-to** iter (pr#21612, songweibin)
  • core: osd/PGLog: get rid of ineffective container operations (pr#19161, xie xingguo)
  • core: osd/PGLog: write only changed dup entries (issue#21026, pr#17245, Josh Durgin)
  • core: osd, pg, mgr: make snap trim queue problems visible (issue#22448, pr#19520, Piotr Dałek)
  • core: osd/PG: misc cleanups (pr#18340, Yan Jun)
  • core: osd/PG: miscellaneous choose acting changes and cleanups (pr#18481, xie xingguo)
  • core: osd/PG: pass scrub priority to replica (pr#20317, Sage Weil)
  • core: osd/PG: perfer async_recovery_targets in reverse order of cost (pr#21578, xie xingguo)
  • core: osd/PG: perfer EC async_recovery_targets in reverse order of cost (pr#21588, xie xingguo)
  • core: osd/PG: PGPool::update: avoid expensive union_of (pr#17239, Zac Medico)
  • core: osd/PGPool::update: optimize with subset_of (pr#17820, Zac Medico)
  • core: osd/PG: reduce some overhead on operating MissingLoc (pr#18186, xie xingguo)
  • core: osd/PG: remote recovery preemption, and new feature bit to condition it on (pr#18553, Sage Weil)
  • core: osd/PG: remove unused parameter in calc_ec_acting (pr#17304, yang.wang)
  • core: osd/PG: restart recovery if NotRecovering and unfound found (issue#22145, pr#18974, Sage Weil)
  • core: osd/PG: revert approx size (issue#22654, pr#18755, Adam Kupczyk)
  • core: osd/PG: re-write of _update_calc_stats and improve pg degraded state (issue#20059, pr#19850, David Zafman)
  • core: osd/PG: some cleanups && add should_gather filter for loop logging (pr#19546, Enming Zhang)
  • core: osd/PG: two cleanups (pr#17171, xie xingguo)
  • core: osd/PG: use osd_backfill_retry_interval for schedule_backfill_retry() (pr#18686, xie xingguo)
  • core: osd/PrimaryLogPG: add condition “is_chunky_scrub_active” to check object in chunky_scrub (pr#18506, Jianpeng Ma)
  • core: osd/PrimaryLogPG: arrange recovery order by number of missing objects (pr#18292, xie xingguo)
  • core: osd/PrimaryLogPG: avoid infinite loop when flush collides with write lock (pr#21653, Sage Weil)
  • core: osd/PrimaryLogPG: calc clone_overlap size in a more efficient and concise way (pr#17928, xie xingguo)
  • core: osd/PrimaryLogPG: cleanup do_sub_op && do_sub_op_reply and define soid in prepare_transaction more appropriate (pr#19495, Enming Zhang)
  • core: osd/PrimaryLogPG: clear data digest on WRITEFULL if skip_data_digest (pr#21676, Sage Weil)
  • core: osd/PrimaryLogPG: clear pin_stats_invalid bit properly on scrub-repair completion (pr#18052, xie xingguo)
  • core: osd/PrimaryLogPG: defer evict if head *or* object intersect scrub interval (issue#23646, pr#21628, Sage Weil)
  • core: osd/PrimaryLogPG: do not pull-up snapc to snapset (pr#18713, Sage Weil)
  • core: osd/PrimaryLogPG: do not set data digest for bluestore (pr#17515, xie xingguo)
  • core: osd/PrimaryLogPG: do not set data/omap digest blindly (pr#18061, xie xingguo)
  • core: osd/PrimaryLogPG: do not use approx_size() for log trimming (pr#18338, xie xingguo)
  • core: osd/PrimaryLogPG: do_osd_ops - propagate EAGAIN/EINPROGRESS on failok (pr#17222, xie xingguo)
  • core: osd/PrimaryLogPG: drop unused parameters (pr#18581, Liao Weizhong)
  • core: osd/PrimaryLogPG: fix dup stat for async read (pr#18693, Xinze Chi)
  • core: osd/PrimaryLogPG: Fix log messages (pr#21639, Gu Zhongyan)
  • core: osd/PrimaryLogPG: fix sparse read won’t trigger repair correctly (pr#17221, xie xingguo)
  • core: osd/PrimaryLogPG: fix the oi size mismatch with real object size (issue#23701, pr#21408, Peng Xie)
  • core: osd/PrimaryLogPG: kick off recovery on backoffing a degraded object (pr#17987, xie xingguo)
  • core: osd/PrimaryLogPG: kill add_interval_usage (pr#17807, xie xingguo)
  • core: osd/PrimaryLogPG: maybe_handle_manifest_detail - sanity check obc existence (pr#17298, xie xingguo)
  • core: osd/PrimaryLogPG: misc cleanups (pr#17830, Yan Jun)
  • core: osd/PrimaryLogPG: more oi.extents fixes (pr#18616, xie xingguo)
  • core: osd/PrimaryLogPG: prepare_transaction - fix EDQUOT vs ENOSPC (pr#17808, xie xingguo)
  • core: osd/PrimaryLogPG: request osdmap update in the right block (issue#21428, pr#17828, Josh Durgin)
  • core: osd/PrimaryLogPG: several oi.extents fixes (pr#18527, xie xingguo)
  • core: osd/PrimaryLogPG: trigger auto-repair on full-object-size CRC error (pr#18353, xie xingguo)
  • core: osd/ReplicatedBackend: clean up code (pr#20127, Jianpeng Ma)
  • core: osd/ReplicatedBackend: ‘osd_deep_scrub_keys’ doesn’t work (pr#20221, fang yuxiang)
  • core: osd/ReplicatedPG: add omap write bytes to delta_stats (pr#18471, Haomai Wang)
  • core: osd_types.cc: reorder fields in serialized pg_stat_t (pr#19965, Piotr Dałek)
  • core: os/filestore: disable rocksdb compression (pr#18707, Sage Weil)
  • core: os/filestore/FileStore: Initialized by nullptr, NULL or 0 instead (pr#18980, Shinobu Kinjo)
  • core: os/filestore: fix device/partition metadata detection (issue#20944, pr#16913, Sage Weil)
  • core: os/filestore: fix do_copy_range replay bug (issue#23298, pr#20832, Sage Weil)
  • core: os/Filestore: fix wbthrottle assert (pr#14213, Xiaoxi Chen)
  • core: os/filestore: print out the error if do_read_entry() fails (pr#18346, Kefu Chai)
  • core: os: FileStore, Using stl min | max, MIN | MAX macros instead (pr#19832, Shinobu Kinjo)
  • core: os: fix 0-length zero semantics, add tests (issue#21712, pr#18159, Sage Weil)
  • core: os/FuseStore: fix incorrect used space statistics for fuse’s statfs interface (pr#19033, Zhi Zhang)
  • core: os/kstore: Drop unused function declaration (pr#18077, Jos Collin)
  • core: os/kstore: fix statfs problem and add vstart.sh support (issue#23590, pr#21287, Yang Honggang)
  • core: os/memstore: Fix wrong use of lock_guard (pr#20914, Shen-Ta Hsieh)
  • core: os/ObjectStore: fix get_data_alignment return -1 causing problem in msg header (pr#18475, Haomai Wang)
  • core: os/ObjectStore.h: fix mistake in comment TRANSACTION ISOLATION (pr#16840, mychoxin)
  • core: os,osd: initial work to drop onreadable/onapplied callbacks (issue#23029, pr#20177, Sage Weil)
  • core: os: unify Sequencer and CollectionHandle (pr#20173, Sage Weil)
  • core: PG: fix name of WaitActingChange (pr#18768, wumingqiao)
  • core: pg: handle MNotifyRec event in down state (pr#20959, Mingxin Liu)
  • core: PGPool::update: optimize removed_snaps comparison when possible (pr#17410, Zac Medico)
  • core: PGPool::update: optimize with interval_set.swap (pr#17121, Zac Medico)
  • core: PG: primary should not be in the peer_info, skip if it is (pr#20189, Neha Ojha)
  • core: ptl-tool: checkout branch after creation (pr#18157, Patrick Donnelly)
  • core: ptl-tool: load labeled PRs (pr#18231, Patrick Donnelly)
  • core: ptl-tool: make branch name configurable (pr#18499, Patrick Donnelly)
  • core: ptl-tool: print bzs/tickets cited in commit msgs/comments (pr#18547, Patrick Donnelly)
  • core: pybind/ceph_argparse: fix cli output info (pr#17667, Luo Kexue)
  • core: pybind/ceph_argparse: Fix UnboundLocalError if command doesn’t validate (pr#21342, Tim Serong)
  • core: pybind/ceph_argparse.py:’timeout’ must in kwargs when call run_in_thread (pr#21659, yangdeliu)
  • core,pybind: pybind/ceph_argparse: accept flexible req (pr#20791, Gu Zhongyan)
  • core,pybind: pybind/rados: add alignment getter to IoCtx (pr#21222, Bruce Flynn)
  • core,pybind: pybind/rados: add rados_service_*() (pr#18812, Kefu Chai)
  • core,pybind: pybind/rados: add support open_ioctx2 API (pr#17710, songweibin)
  • core,pybind: rados: support python API of “set_osdmap_full_try” (pr#17418, songweibin)
  • core: qa: fix the potential delay of pg state change (pr#17253, huangjun)
  • core: qa/standalone/osd/repro_long_log: no-mon-config for cot (pr#20919, Sage Weil)
  • core: qa/standalone/scrub/osd-scrub-repair: no -y to diff (issue#21618, pr#18079, Sage Weil)
  • core: qa/suite/rados: fix balancer vs firefly tunables failures (pr#18826, Sage Weil)
  • core: qa/suites/rados: fewer msgr failures (pr#20918, Sage Weil)
  • core: qa/suites/rados/perf: whitelist health warnings (pr#18878, Sage Weil)
  • core: qa/suites/rados/rest/mgr: provision openstack volumes (pr#20573, Sage Weil)
  • core: qa/suites/rados/singleton/all/mon-seesaw: whitelist MON_DOWN (pr#18246, Sage Weil)
  • core: qa/suites/rados/singleton/all/recover-preemption: handle slow starting osd (pr#18078, Sage Weil)
  • core: qa/suites/rados/singleton/all/recovery_preemption: whitelist SLOW_OPS (pr#21250, Sage Weil)
  • core: qa/suites/rados/singleton/diverget_priors*: broaden whitelist (pr#17379, Sage Weil)
  • core: qa/suites/rados/thrash: extend mgr beacon grace when many msgr failures injected (issue#21147, pr#19242, Sage Weil)
  • core: qa/suites/rados/verify/tasks/rados_api_tests: whitelist OBJECT_MISPLACED (pr#21646, Sage Weil)
  • core: qa/workunits/rest/test.py: stop trying to test obsolte cluster_up/down (pr#18552, Sage Weil)
  • core: rados/objclass.h: fix build define CEPH_CLS_API in all cases (pr#21606, Danny Al-Gaaf)
  • core: rados: use WaitInterval()’s return value instead of manual timing (pr#20028, Mohamad Gebai)
  • core,rbd: common,rbd-nbd: fix up prefork behavior vs AsyncMessenger singletons (issue#23143, pr#20681, Sage Weil)
  • core,rbd: librbd,os: address coverity false positives (pr#17793, Amit Kumar)
  • core,rbd: mgr,osd,kv: Fix various warnings for Clang and GCC7 (pr#17976, Adam C. Emerson)
  • core,rbd: vstart.sh: fix mstart variables (pr#20826, Sage Weil)
  • core: rdma: Assign instead of compare (pr#16664, amitkuma)
  • core: remove startsync (issue#20604, pr#16396, Amit Kumar)
  • core: rocksdb: sync with upstream (issue#20529, pr#17388, Kefu Chai)
  • core: rocksdb: sync with upstream (pr#21320, Kefu Chai)
  • core: scrub errors not cleared on replicas can cause inconsistent pg state when replica takes over primary (issue#23267, pr#21101, David Zafman)
  • core: Snapset inconsistency is detected with its own error (issue#22996, pr#20450, David Zafman)
  • core: src/messages/MOSDMap: reencode OSDMap for older clients (issue#21660, pr#18134, Sage Weil)
  • core: src/osd/PG.cc: 6455: FAILED assert(0 == “we got a bad state machine event”) (pr#20933, David Zafman)
  • core: src/test/osd: add two pool test for manifest objects (pr#20096, Myoungwon Oh)
  • core: test/cli/osdmaptool/test-map-pgs.t: remove nondetermimistic test (pr#20872, Sage Weil)
  • core: test/objectstore_bench: Don’t forget judging whether call usage (pr#21369, Jianpeng Ma)
  • core,tests: ceph_test_filestore_idempotent_sequence: many fixes (issue#22920, pr#20279, Sage Weil)
  • core,tests: ceph_test_objectstore: drop expect regex (pr#16968, Sage Weil)
  • core,tests: Erasure code read test and code cleanup (issue#14513, pr#17703, David Zafman)
  • core,tests: Erasure code recovery should send additional reads if necessary (issue#21382, pr#17920, David Zafman)
  • core,tests: osd,dmclock: fix dmclock test simulator change (pr#20270, J. Eric Ivancich)
  • core,tests: os: kstore fix unittest for FiemapHole (pr#17313, Ning Yao)
  • core,tests: os/memstore: memstore_page_set=false (issue#20738, pr#16995, Sage Weil)
  • core,tests: qa/ceph_manager: check pg state again before timedout (issue#21294, pr#17810, huangjun)
  • core,tests: qa/clusters/fixed-[23]: 4 osds per node, not 3 (pr#16799, Sage Weil)
  • core,tests: qa: modify rgw default pool names (pr#21630, Neha Ojha)
  • core,tests: qa/objectstore/bluestore*: less debug output (issue#20910, pr#17505, Sage Weil)
  • core,tests: qa/overrides/2-size-2-min-size: whitelist REQUEST_STUCK (pr#17243, Sage Weil)
  • core,tests: qa/standalone/ceph-helpers: pass –verbose to ceph-disk (pr#19456, Sage Weil)
  • core,tests: qa/standalone/scrub/osd-scrub-repair: fix grep pattern (issue#21127, pr#17258, Sage Weil)
  • core,tests: qa/standalone/scrub/osd-scrub-snaps: adjust test for lack of snapdir objects (pr#17927, Sage Weil)
  • core,tests: qa/suites/rados/monthrash: tolerate PG_AVAILABILITY during mon thrashing (pr#18122, Sage Weil)
  • core,tests: qa/suites/rados/monthrash: whitelist SLOW_OPS (pr#21331, Sage Weil)
  • core,tests: qa/suites/rados/objectstore: logs (issue#20738, pr#16923, Sage Weil)
  • core,tests: qa/suites/rados/perf: create pool with lower pg_num (pr#17819, Neha Ojha)
  • core,tests: qa/suites/rados/rest/mgr-restful: whitelist more health (pr#18457, Sage Weil)
  • core,tests: qa/suites/rados/rest: move rest_test from qa/suites/rest/ (pr#19175, Sage Weil)
  • core,tests: qa/suites/rados/thrash: fix thrashing with ec vs map discon (pr#16842, Sage Weil)
  • core,tests: qa/suites/rados/thrash-old-clients: add hammer clients (pr#21703, Sage Weil)
  • core,tests: qa/suites/rados/thrash-old-clients: add rbd tests (pr#21704, Sage Weil)
  • core,tests: qa/suites/rados/thrash-old-clients: do some thrashing with jewel and luminous clients (pr#21679, Sage Weil)
  • core,tests: qa/suites/rados/thrash-old-clients: only centos and 16.04 (pr#22125, Sage Weil)
  • core,tests: qa/suites/upgrade/jewel-x/stress-split: tolerate sloppy past_intervals (pr#17226, Sage Weil)
  • core,tests: qa/suites/upgrade/luminous-x/stress-split: avoid enospc (pr#21753, Sage Weil)
  • core,tests: qa/tasks/ceph_manager: revive osds before doing final rerr reset (issue#21206, pr#17406, Sage Weil)
  • core,tests: qa/tasks/ceph_manager: tolerate tell osd.* error (pr#19365, Sage Weil)
  • core,tests: qa/tasks/ceph.py: tolerate flush pg stats exception (pr#16905, Sage Weil)
  • core,tests: qa/tasks/filestore_idempotent: shorter test (pr#20509, Sage Weil)
  • core,tests: qa/tasks/thrashosds: set min_in default to 4 (issue#21997, pr#18670, Sage Weil)
  • core,tests: qa/tests: run ceph-ansible task on installer.0 role/node (pr#19605, Yuri Weinstein)
  • core,tests: qa: tolerate failure to force backfill (issue#22614, pr#19765, Sage Weil)
  • core,tests: qa/workunits/rados/test_rados_tool: fix stray |, race (issue#22676, pr#19946, Sage Weil)
  • core,tests: qa/workunits/rados/test.sh: ensure tee output is valid filename (pr#21507, Sage Weil)
  • core,tests: rados: Initialization of alignment (pr#17723, Amit Kumar)
  • core,tests: rados: Initializing members of librados/TestCase.h (pr#16896, amitkuma)
  • core,tests: test: Checking fd for negative before closing (pr#17190, amitkuma)
  • core,tests: test: Check to avoid divide by zero (pr#17220, amitkuma)
  • core: tool: change default objectstore from filestore to bluestore (pr#18066, Song Shun)
  • core: tool: misc cleanup of ceph-kvstore-tool (issue#22092, pr#18815, Chang Liu)
  • core,tools: Add export and remove ceph-objectstore-tool command option (issue#21272, pr#17538, David Zafman)
  • core,tools: ceph-objectstore-tool: fix import of post-split pg from pre-split ancestor (issue#21753, pr#18229, Sage Weil)
  • core: tools/ceph-objectstore-tool: split filestore directories offline to target hash level (issue#21366, pr#17666, Zhi Zhang)
  • core,tools: common, tool: update kvstore-tool to repair key/value database (issue#17730, issue#21744, pr#16745, liuchang0812, Chang Liu)
  • core,tools: osd,os/bluestore: kill clang analyzer warnings (pr#18015, Kefu Chai)
  • core: tools/rados: add touch command to change object modification time (pr#18913, Yao Zongyou)
  • core,tools: scripts: add ptl-tool for scripting merges (pr#17926, Patrick Donnelly)
  • core: vstart.sh: drop .ceph_port and use randomly selected available port (pr#19268, Shinobu Kinjo)
  • core: vstart.sh: drop –{mon,osd,mds,rgw,mgr}_num options (pr#18648, Kefu Chai)
  • core: vstart.sh: Remove duplicate global section (pr#17543, iliul)
  • crush: cleanup update_device_class() log messages (pr#21174, Gu Zhongyan)
  • crush: fix CrushCompiler won’t compile maps with empty shadow tree (pr#17058, xie xingguo)
  • crush: fix device_class_clone for unpopulated/empty weight-sets (issue#23386, pr#22169, Sage Weil)
  • crush: fix fast rule lookup when uniform (pr#17510, Sage Weil)
  • crush: force rebuilding shadow hierarchy after swapping buckets (pr#17083, xie xingguo)
  • crush: improve straw2 algorithm’s readability (pr#20196, Yao Zongyou)
  • crush: “osd crush class rename” support (pr#16961, xie xingguo)
  • crush, osd: handle multiple parents properly when applying pg upmaps (issue#23921, pr#21835, xiexingguo)
  • crush: safe check for ‘ceph osd crush swap-bucket’ (pr#17335, Carudy)
  • crush: various CrushWrapper cleanups (pr#17360, Kefu Chai)
  • crush: various weight-set fixes (pr#17014, xie xingguo)
  • denc: should check element’s type not ‘size_t’ (pr#19986, Kefu Chai)
  • denc: use constexpr-if to replace some SFINAE impls (pr#19662, Kefu Chai)
  • doc: 12.1.3 release notes (pr#16975, Abhishek Lekshmanan)
  • doc: 12.2.0 major release announcements (pr#16915, Abhishek Lekshmanan)
  • doc: 12.2.1 release notes (pr#18014, Abhishek Lekshmanan)
  • doc: 12.2.4 release notes (pr#20619, Abhishek Lekshmanan)
  • doc: add 12.2.2 release notes (pr#19264, Abhishek Lekshmanan)
  • doc: add allow_multimds and fs_name parameter (pr#15847, Jan Fajerski)
  • doc: add ceph-kvstore-tool’s man (pr#17092, liuchang0812)
  • doc: add changelog for 12.2.1 (pr#18020, Abhishek Lekshmanan)
  • doc: add changelog for v11.2.1 (pr#16956, Abhishek Lekshmanan)
  • doc: add changelog for v12.2.2 (pr#19284, Abhishek Lekshmanan)
  • doc: Added CHAP configuration instructions for iSCSI (pr#18423, Ashish Singh)
  • doc: add example of setting pool in cephfs layout (pr#17372, John Spray)
  • doc: Adding changelog for 10.2.10 (pr#18151, Abhishek Lekshmanan)
  • doc: Add introduction about different way to run rbd-mirror (pr#19692, Yu Shengzuo)
  • doc: add –max-buckets to radosgw-admin(8) (pr#17439, Clément Pellegrini)
  • doc: add missing blank line (pr#18724, iliul)
  • doc: Add missing pg states from doc (pr#20504, David Zafman)
  • doc: add mount.fuse.ceph to index (issue#22595, pr#19792, Jos Collin)
  • doc: Add newbie-friendly updates to Helm start doc (pr#18886, Blaine Gardner)
  • doc: add osd_max_object_size in osd configuration (pr#18115, Mohamad Gebai)
  • doc: build-doc: Upgrade ceph python libraries (pr#20726, Boris Ranto)
  • doc: ceph-disk: create deprecation warnings (issue#22154, pr#18988, Alfredo Deza)
  • doc: ceph-volume: automatic VDO detection (issue#23581, pr#21451, Alfredo Deza)
  • doc: ceph-volume docs (pr#17068, Alfredo Deza)
  • doc: ceph-volume document multipath support (pr#20878, Alfredo Deza)
  • doc: ceph-volume doc updates (pr#20758, Alfredo Deza)
  • doc: ceph-volume include physical devices associated with an LV when listing (pr#21645, Alfredo Deza)
  • doc: ceph-volume lvm bluestore support (pr#18448, Alfredo Deza)
  • doc/ceph-volume OSD use the fsid file, not the osd_fsid (issue#22427, pr#20059, Alfredo Deza)
  • doc: change boolean option default value from zero to false (pr#17733, Yao Zongyou)
  • doc: change cn mirror to ustc domain (pr#18081, Shengjing Zhu)
  • doc: changelog for v12.2.3 (pr#20503, Abhishek Lekshmanan)
  • doc: cleanup erasure coded pool doc on cephfs use (pr#20572, Patrick Donnelly)
  • doc: CodingStyle: add python and javascript/typescript (pr#20186, Joao Eduardo Luis)
  • doc: common/options: document filestore and filejournal options (pr#17739, Sage Weil)
  • doc: common/options: document objecter, filer, and journal options (pr#17740, Sage Weil)
  • doc: complete and update the subsystem logging level info table (pr#18500, Luo Kexue)
  • doc: correcting typos in bluestore-config-ref and bluestore-migration (pr#19154, Katie Holly)
  • doc: correct wrong bluestore config types (pr#18205, Yao Zongyou)
  • doc: delete duplicate words (pr#17104, iliul)
  • doc: dev description of async recovery (pr#21051, Neha Ojha, Josh Durgin)
  • doc: doc/bluestore: add SPDK usage for bluestore (pr#17654, Haomai Wang)
  • doc: doc/cephfs/experimental-features: kernel client snapshots limit (pr#18579, Ilya Dryomov)
  • doc: doc/cephfs/posix: remove stale information for seekdir (pr#17658, “Yan, Zheng”)
  • doc: doc/conf.py: do not set html_use_smartypants explicitly (pr#17127, Kefu Chai)
  • doc: doc/dev: add a brief guide to serialization (pr#20131, John Spray)
  • doc: doc/dev/cxx: add C++11 ABI related doc (pr#20030, Kefu Chai)
  • doc: doc/dev/iana: document our official IANA numbers (pr#16910, Sage Weil)
  • doc: doc/dev/index: update rados lead (pr#16911, Sage Weil)
  • doc: doc/dev/macos: add doc for building on MacOS (pr#20031, Kefu Chai)
  • doc: doc/dev/msgr2.rst: a few notes on protocol goals (pr#20083, Sage Weil)
  • doc: doc/dev/perf: add doc on disabling -fomit-frame-pointer (pr#17358, Kefu Chai)
  • doc: doc for mount.fuse.ceph (issue#21539, pr#19172, Jos Collin)
  • doc: doc/man remove deprecation of ceph-disk man page title (pr#19325, Alfredo Deza)
  • doc: doc/mgr: Add limitations section to plugin guide (pr#21347, Tim Serong)
  • doc: doc/mgr: add “local pool” plugin to toc (pr#17961, Kefu Chai)
  • doc: doc/mgr/balancer: document (issue#22789, pr#21421, Sage Weil)
  • doc: doc/mgr: document facilities methods using automethod directive (pr#18680, Kefu Chai)
  • doc: doc/mgr/plugins: add note about distinction between config and kv store (pr#21671, Jan Fajerski)
  • doc: doc/mgr: remove non user-facing code from doc (pr#20372, Kefu Chai)
  • doc: doc,os,osdc: drop and modify comments (pr#17732, songweibin)
  • doc: doc/rados: Add explanation of straw2 (pr#19247, Shinobu Kinjo)
  • doc: doc/rados/operations/bluestore-migration: document bluestore migration process (pr#16918, Sage Weil)
  • doc: doc/rados/operations/bluestore-migration: update docs a bit (pr#17011, Sage Weil)
  • doc: doc raise exceptions with a base class (pr#18152, Alfredo Deza)
  • doc: doc/rbd: add info for rbd group (pr#17633, yonghengdexin735)
  • doc: doc/rbd: add missing several commands in rbd CLI man page (issue#14539, issue#16999, pr#19659, songweibin)
  • doc: doc/rbd: correct the path of librbd python APIs (pr#19690, songweibin)
  • doc: doc/rbd: fix typo s/morror/mirror (pr#19997, songweibin)
  • doc: doc/rbd: iSCSI Gateway Documentation (issue#20437, pr#17376, Aron Gunn, Jason Dillaman)
  • doc: doc/rbd: specify additional ESX prerequisites (pr#18517, Jason Dillaman)
  • doc: doc/rbd: tweaks for the LIO iSCSI gateway (issue#21763, pr#18250, Jason Dillaman)
  • doc: doc/rbd: tweaks to the Windows iSCSI initiator directions (pr#18704, Jason Dillaman)
  • doc: doc/release-notes: add jewel->kraken notes (pr#18482, Sage Weil)
  • doc: doc/release-notes: clarify purpose of require-osd-release (pr#17270, Sage Weil)
  • doc: doc/release-notes: clarify that you need to keep your existing OSD caps (pr#18825, Jason Dillaman)
  • doc: doc/release-notes: ensure RBD users can blacklist prior to upgrade (issue#21353, pr#17755, Jason Dillaman)
  • doc: doc/release-notes: fix typo ‘psd’ to ‘osd’ (pr#18695, wangsongbo)
  • doc: doc/releases: the Kraken sleepeth, faintest sunlights flee (pr#17424, Abhishek Lekshmanan)
  • doc: doc/releases: update release cycle docs (pr#18117, Sage Weil)
  • doc: doc/rgw: add page for http frontend configuration (issue#13523, pr#20058, Casey Bodley)
  • doc: doc/scripts: py3 compatible (pr#17640, Kefu Chai)
  • doc: docs: Do not use “min size = 1” as an example (pr#17912, Alfredo Deza)
  • doc: docs fix ceph-volume missing sub-commands (issue#23148, pr#20673, Alfredo Deza)
  • doc: doc/start/os-recommendations.rst: bump krbd kernels (pr#21478, Ilya Dryomov)
  • doc: docs update ceph-deploy reference to reflect ceph-volume API (pr#20510, Alfredo Deza)
  • doc: document ceph-disk prepare class hierarchy (pr#17019, Loic Dachary)
  • doc: document include/ipaddr.h (issue#12056, pr#17613, Nathan Cutler)
  • doc: drop duplicate line in ceph-bluestore-tool man page (pr#19169, Xiaojun Liao)
  • doc: eliminate useless cat statement (pr#17154, Ken Dreyer)
  • doc: examples: add new librbd example (pr#18314, Mahati Chamarthy)
  • doc: expand developer documentation of unit tests (pr#19594, Nathan Cutler)
  • doc: Fix a grammar error in rbd-snapshot.rst (pr#21470, Zeqing Tyler Qi)
  • doc: fix CFLAGS in doc/dev/cpu-profiler.rst (pr#19752, Chang Liu)
  • doc: fix desc of option “mon cluster log file” (pr#18770, Kefu Chai)
  • doc: fix doc/radosgw/admin.rst typos (pr#17397, Enming Zhang)
  • doc: Fix dynamic resharding doc formatting (pr#20970, Ashish Singh)
  • doc: fix error in osd scrub load threshold (pr#21678, Dirk Sarpe)
  • doc: Fixes a spelling error and a broken hyperlink (pr#20442, Jordan Hus)
  • doc: Fixes rbd snapshot flatten example (issue#17723, pr#17436, Ashish Singh)
  • doc: fixes syntax in osd-config-ref (issue#21733, pr#18188, Joshua Schmid)
  • doc: Fixes the name of the CephFS snapshot directory (pr#18710, Jordan Rodgers)
  • doc: fix hyper link to radosgw/config-ref (pr#17986, Kefu Chai)
  • doc: fix librbdpy example (pr#20019, Yuan Zhou)
  • doc: fix order of options in osd new (issue#21023, pr#17326, Neha Ojha)
  • doc: fix sphinx build warnings and errors (pr#17025, Alfredo Deza)
  • doc: fix the desc of “osd max pg per osd hard ratio” (pr#18373, Kefu Chai)
  • doc: Fix typo and URL (pr#18040, Jos Collin)
  • doc: fix typo e.g,. => e.g (pr#18607, Yao Zongyou)
  • doc: fix typo in bluestore-migration.rst (pr#18389, Yao Zongyou)
  • doc: Fix typo in mount.fuse.ceph (pr#19215, Jos Collin)
  • doc: fix typo in php.rst (pr#17762, Yao Zongyou)
  • doc: fix typo in radosgw/dynamicresharding.rst (pr#18651, Alexander Ermolaev)
  • doc: fix typo on specify db block device (pr#17590, Xiaoxi Chen)
  • doc: Fix typo s/applicatoin/application/ (pr#20720, Francois Deppierraz)
  • doc: Fix typos in placement-groups.rst (pr#17973, Matt Boyle)
  • doc: Fix typos in release notes (pr#18950, Stefan Knorr)
  • doc: .githubmap: Add cbodley (pr#18946, Jos Collin)
  • doc: githubmap: add map for GitHub contributor lookup (pr#17457, Patrick Donnelly)
  • doc: .githubmap, .mailmap, .organizationmap, .peoplemap: update Igor (pr#19314, Igor Fedotov)
  • doc: globally change CRUSH ruleset to CRUSH rule (issue#20559, pr#19435, Nathan Cutler)
  • doc: Improved dashboard documentation (pr#21443, Lenz Grimmer)
  • doc: Improved hitset parameters description (pr#19691, Alexey Stupnikov)
  • doc: improve links in doc/releases.rst (pr#18155, Nathan Cutler)
  • doc: Improve mgr/restful module documentation (pr#20717, Boris Ranto)
  • doc: Improve the ceph fs set max_mds command (issue#21007, pr#17044, Bara Ancincova)
  • doc: include ceph-disk and ceph-disk-volume man pages in index (pr#17168, Alfredo Deza)
  • doc: init flags to 0 in rados example (pr#20671, Patrick Donnelly)
  • doc: Kube + Helm installation (pr#18520, Alexandre Marangone)
  • doc: legal: remove doc license ambiguity (issue#23336, pr#20876, Nathan Cutler)
  • doc: lock_timeout is a per mapping option (pr#21563, Ilya Dryomov)
  • doc: log-and-debug: fix default value of “log max recent” (pr#20316, Nathan Cutler)
  • doc: mailmap: Add Sibei, XueYu Affiliation (pr#18395, Sibei Gao)
  • doc: mailmap: Fixed maintenance guide URL (pr#18076, Jos Collin)
  • doc: mailmap, organizationmap: add Dongsheng, Liuzhong, Pengcheng, Yang Affiliation (pr#17548, Dongsheng Yang)
  • doc: .mailmap, .organizationmap: add Fufei, Mingqiao and Ying Affiliation (pr#17540, Ying He)
  • doc: .mailmap, .organizationmap: Add Liu Lei’s mailmap and affiliation (pr#17105, iliul)
  • doc: .mailmap, .organizationmap: update JingChen, ZongyouYao, ShanchunLv’s… (pr#18960, Chang Liu)
  • doc: mailmap: update affiliation for Mykola Golub (pr#18069, Mykola Golub)
  • doc: mailmap: update affiliation for Mykola Golub (pr#19667, Mykola Golub)
  • doc: mailmap: Update umcloud affiliation (pr#17441, Yixing Yan)
  • doc: make the commands in README.md properly aligned (pr#18639, Yao Zongyou)
  • doc/man: add “ls” to “ceph osd” command’s subcommands list (pr#19382, Rishabh Dave)
  • doc: “mds blacklist interval” vs manually blacklisting (pr#18195, Ken Dreyer)
  • doc: mgr/dashboard.rst: mention ceph.conf and ceph mgr services (pr#20961, Nathan Cutler)
  • doc/mgr/plugins: mgr accessor during init causes exception (pr#16973, Jan Fajerski)
  • doc: mimic: doc: Updated dashboard documentation (features, SSL config) (pr#22079, Lenz Grimmer)
  • doc: misc fix spell errors in osd/OSD and doc (pr#17107, songweibin)
  • doc: misc: fix various spelling errors (pr#20831, Shengjing Zhu)
  • doc: Misc iSCSI doc updates (pr#19931, Mike Christie)
  • doc: move glance_api_version option to the right place (pr#17337, Luo Kexue)
  • doc: options.cc: document rgw config options (pr#18007, Yehuda Sadeh)
  • doc: organizationmap: Add Adam Wolfe Gordon’s affiliation (pr#18295, Adam Wolfe Gordon)
  • doc: organizationmap: Add Ashish Singh affiliation (pr#17109, Ashish Singh)
  • doc: .organizationmap: add Xin Yuan and Yichao Li’s affiliation (pr#21170, Li Wang)
  • doc: PendingReleaseNotes: Added note about Dashboard v2, fixed typo (pr#21597, Lenz Grimmer)
  • doc: PendingReleaseNotes:Announce FreeBSD availability (pr#16782, Willem Jan Withagen)
  • doc: PendingReleaseNotes: mention some monitor changes (pr#21474, Joao Eduardo Luis)
  • doc: PendingReleaseNotes: note about upmap mapping change in luminous release notes (pr#17813, Sage Weil)
  • doc: qa,doc: drop support of ubuntu trusty (pr#19307, Kefu Chai)
  • doc/rados/operations/bluestore-migration: typos and whitespace (pr#16991, Sage Weil)
  • doc/rados/operations/bluestore-migration: typos (pr#17581, Sage Weil)
  • doc: README: Improve vstart.sh usage (pr#17644, Fabian Vogt)
  • doc: README.md: bump up cmake to 2.8.12 (pr#18348, Yan Jun)
  • doc: redundant “cephfs” when set the “allow_multimds” (pr#20045, Shangzhong Zhu)
  • doc: release notes: fix grammar/style nits (pr#18876, Nathan Cutler)
  • doc: release notes for 12.2.3 (pr#20500, Abhishek Lekshmanan)
  • doc: release notes for v12.1.4 Luminous (pr#17037, Abhishek Lekshmanan)
  • doc/release-notes: remove mention of crush weight optimization (pr#16974, Sage Weil)
  • doc: release-notes.rst: add Kraken v11.2.1 and update releases.rst (pr#16879, Nathan Cutler)
  • doc: release notes update for 10.2.10 (pr#18148, Abhishek Lekshmanan)
  • doc/releases: drop LTS/stable line from second chart (pr#18153, Sage Weil)
  • doc: Remove additional arguments when replacing OSD (pr#18345, Wido den Hollander)
  • doc: remove duplicated –max-buckets option desc (pr#19737, Kefu Chai)
  • doc: remove references to unversioned repository addresses (pr#21357, Greg Farnum)
  • doc: remove unused config: “osd op threads” (pr#21319, Jianpeng Ma)
  • doc: rename changelog with a .txt extension (pr#18156, Abhishek Lekshmanan)
  • doc: reorganize releases (pr#20784, Abhishek Lekshmanan)
  • doc: replace injectargs usage with “config set” (pr#18789, John Spray)
  • doc: replace region with zonegroup in configure bucket sharding section (issue#21610, pr#18063, Orit Wasserman)
  • doc: restructure bluestore migration insructions (pr#17603, Sage Weil)
  • doc: Revise the Example of Bucket Policy (pr#17362, zhangwen)
  • doc: rgw: add a note for resharding in 12.2.1 docs (pr#17675, Abhishek Lekshmanan)
  • doc: rgw add some basic documentation for sync plugins & ES (pr#15849, Abhishek Lekshmanan)
  • doc: rgw adminops binding libraries (pr#19164, hrchu)
  • doc: rgw mention about tagging & bucket policies in s3api (pr#16907, Abhishek Lekshmanan)
  • doc: rgw: mention the civetweb support for binding to multiple ports (issue#20942, pr#17141, Abhishek Lekshmanan)
  • doc: rm stray “)” character from mds config ref (pr#18228, Ken Dreyer)
  • docs: ceph-volume CLI updates (pr#17425, Alfredo Deza)
  • doc: s/deamon/daemon/ (pr#20931, ashitakasam)
  • doc: some improvements to ceph-conf.rst (pr#21268, Nathan Cutler)
  • doc: Specify mount details in ceph-fuse (pr#20071, Jos Collin)
  • doc: SubmittingPatches: clarify PR title section (pr#17143, Nathan Cutler)
  • doc/templates update toctree call to include hidden entries (pr#17076, Alfredo Deza)
  • doc: the client inputs the pool name instead of pool ID (pr#17672, Frank Yu)
  • doc: typo fix (pr#21077, Ashita Dashottar)
  • doc: update Blacklisting and OSD epoch barrier (issue#22542, pr#19701, Jos Collin)
  • doc: update ceph-disk with a state-transition diagram (pr#17639, Kefu Chai)
  • doc: update ceph iscsi kernel and package info (pr#20020, Mike Christie)
  • doc: Update commands and options in radosgw-admin (pr#18267, Jos Collin)
  • doc: update Component Technical Leads and maintainers to canonical location (pr#18376, Patrick McGarry)
  • doc: Update config file search paths to reflect reality (pr#19882, Adam Wolfe Gordon)
  • doc: updated add primary storage documentation for latest CloudStack release (4.11) (pr#21050, James McClune, John Wilkins)
  • doc: Update dashboard feature list (added RGW management) (pr#21781, Lenz Grimmer)
  • doc: updated dashboard feature list (added new RGW details, Pools) (pr#21562, Lenz Grimmer)
  • doc: Updated dashboard feature list (pr#21693, Lenz Grimmer)
  • doc: Updated dashboard v2 feature list (pr#20755, Lenz Grimmer)
  • doc: Updated documentation for Zabbix Mgr module (pr#18356, Wido den Hollander)
  • doc: update default value of option mon_sync_timeout (pr#17802, Yao Guotao)
  • doc: update default value of parameter mon_subscribe_interval (pr#17669, yaoguotao)
  • doc: Update docs to remove gitbuilder and add shaman references (pr#17022, Alfredo Deza)
  • doc: updated the dashboard feature list (pr#21531, Lenz Grimmer)
  • doc: Updated the get-packages.rst to luminous (pr#20815, Kai Wagner)
  • doc: update firewall doc to mention ceph-mgr (pr#17974, John Spray)
  • doc: update iSCSI upstream kernel to 4.16 (pr#20695, Mike Christie)
  • doc: update link to placing-different-pools (pr#17833, Mohamad Gebai)
  • doc: update Li Wang Affiliation (pr#18060, Li Wang)
  • doc: update man page to explain ceph-volume support bluestore (issue#22663, pr#19960, lijing)
  • doc: Update manual deployment (issue#20309, pr#15811, Jens Rosenboom)
  • doc: update mgr/dashboard doc about standbys (pr#19879, John Spray)
  • doc: Update mgr doc on how to enable Zabbix module (pr#16861, Wido den Hollander)
  • doc: update mgr related auth settings (pr#20126, Kefu Chai)
  • doc: Update monitoring.rst (pr#20630, Jos Collin)
  • doc: update rbd-mirroring documentation (issue#20701, pr#16908, Jason Dillaman)
  • doc: update references to use ceph-volume (pr#19241, Alfredo Deza)
  • doc: update releases to the current state (pr#17364, Abhishek Lekshmanan)
  • doc: Updates to bluestore migration doc (pr#17602, David Galloway)
  • doc: v12.2.5 luminous release notes (pr#21621, Abhishek Lekshmanan)
  • doc: various cleanups (pr#18480, Kefu Chai)
  • examples: fix link order in librados example Makefile (pr#17842, Mahati Chamarthy)
  • Fix ceph-mgr restarts (pr#22051, Boris Ranto)
  • follow-up fixups for atomic_t spinlocks (pr#17611, Jesse Williamson)
  • githubmap: Add ktdreyer (pr#19209, Jos Collin)
  • include/buffer.h: fix typo in comment (pr#17489, mychoxin)
  • include/ceph_features: fix OS_PERF_STAT_NS’s incarnation (pr#21467, Kefu Chai)
  • install-deps.sh: fix an error condition expression (pr#20819, Yao Guotao)
  • java/native: fix milliseconds to mtime/atime conversion (pr#17460, dengquan)
  • java/native: s/jni: lstat/jni: stat in native_ceph_stat (pr#20142, Shangzhong Zhu)
  • KStore: statfs needs extra includes on FreeBSD (pr#21429, Willem Jan Withagen)
  • kv/leveldb: fix deadlock when close db (pr#16643, Zengran)
  • kv: unify {create_and_,}open() methods (pr#18177, Kefu Chai)
  • librados: add async interfaces for use with Networking TS (pr#19054, Casey Bodley)
  • librados: block MgrClient::start_command until mgrmap (pr#21832, John Spray, Kefu Chai)
  • librados: extend C API for so it accepts keys with NUL chars (pr#20314, Piotr Dałek)
  • librados: Fix a potential risk of buffer::list::claim_prepend(list& b… (issue#21338, pr#17661, Guan yunfei)
  • librados: fix potential race condition if notify immediately fails (issue#23966, pr#21859, Jason Dillaman)
  • librados: getter for min compatible client versions (pr#20080, Jason Dillaman)
  • librados: invalid free() in rados_getxattrs_next() (issue#22042, pr#20260, Gu Zhongyan)
  • librados: make OPERATION_FULL_FORCE the default for rados_remove() (issue#22413, pr#20534, Kefu Chai)
  • librbd: abstract hard-coded journal and cache hooks on IO path (pr#20682, Jason Dillaman)
  • librbd: Add a function to list image watchers (pr#19188, Adam Wolfe Gordon)
  • librbd: add API function to get image name (pr#20935, Mykola Golub)
  • librbd: added preprocessor macro for detecting compare-and-write support (issue#22036, pr#18708, Jason Dillaman)
  • librbd: add eventtrace support (pr#19251, Mahati Chamarthy)
  • librbd: add preliminary support for new operation feature bit (pr#19903, Jason Dillaman)
  • librbd: address coverity false positives (pr#17696, Amit Kumar)
  • librbd: address coverity false positives (pr#17721, Amit Kumar)
  • librbd: auto-remove trash snapshots when image is deleted (issue#22873, pr#20376, Jason Dillaman)
  • librbd: by default use new format for deep copy destination (pr#20222, Mykola Golub)
  • librbd: cache last index position to accelerate snap create/rm (issue#22716, pr#19974, Song Shun)
  • librbd: cannot clone all image-metas if we have more than 64 key/value pairs (pr#18327, PCzhangPC)
  • librbd: cannot copy all image-metas if we have more than 64 key/value pairs (pr#18328, PCzhangPC)
  • librbd: clean up ManagedLock log prefix (pr#20159, shun-s)
  • librbd: compare and write against a clone can result in failure (issue#20789, pr#18887, Jason Dillaman)
  • librbd: deep_copy: don’t create snapshots above snap_id_end (pr#19383, Mykola Golub)
  • librbd: default localize parent reads to false (issue#20941, pr#16882, Jason Dillaman)
  • librbd: default to sparse-reads for any IO operation over 64K (issue#21849, pr#18405, Jason Dillaman)
  • librbd: disable ENOENT tracking within the object cacher (issue#23597, pr#21308, Jason Dillaman)
  • librbd: disallow creation of v1 image format (pr#20460, Julien COLLET, Julien Collet)
  • librbd: don’t read metadata twice on image open (pr#18542, Mykola Golub)
  • librbd: drop redundant check for null ImageCtx (pr#18265, Jianpeng Ma)
  • librbd: filter out potential race with image rename (issue#18435, pr#19618, Jason Dillaman)
  • librbd: fix coverity warning for uninitialized member (pr#18129, Li Wang)
  • librbd: fix deep copy a child-image (pr#20099, songweibin)
  • librbd: fix don’t send get_stripe_unit_count if striping is not enabled (issue#21360, pr#17660, Yanhu Cao)
  • librbd: fix issues discovered in clone v2 during upgrade tests (issue#22979, pr#20406, Jason Dillaman)
  • librbd: fix missing return in NotifyMessage::get_notify_op (pr#20656, Yao Zongyou)
  • librbd: fix rbd close race with rewatch (pr#21141, Song Shun)
  • librbd: fix refuse to release lock when cookie is the same at rewatch (pr#20868, Song Shun)
  • librbd: fix structure size check in rbd_mirror_image_get_info/status (pr#20478, Mykola Golub)
  • librbd: force removal of a snapshot cannot ignore dependent children (issue#22791, pr#20105, Jason Dillaman)
  • librbd: generalized deep copy function (pr#16238, Mykola Golub)
  • librbd: group and snapshot cleanup (pr#19990, Jason Dillaman)
  • librbd: group snapshots (pr#11544, Victor Denisov, Jason Dillaman)
  • librbd: hold cache_lock while clearing cache nonexistence flags (issue#21558, pr#17992, Jason Dillaman)
  • librbd: image-meta config overrides should be dynamically refreshed (issue#21529, pr#18042, Dongsheng Yang, Jason Dillaman)
  • librbd: initial hooks for clone v2 support (pr#20176, Jason Dillaman)
  • librbd: initialization of state member variables (pr#16866, amitkuma)
  • librbd: Initializing members image,operation,journal (pr#16934, amitkuma)
  • librbd: Initializing member variables (pr#16867, amitkuma)
  • librbd: journal should ignore -EILSEQ errors from compare-and-write (issue#21628, pr#18099, Jason Dillaman)
  • librbd,librados: do not include stdbool.h in C++ headers (pr#19945, Kefu Chai)
  • librbd: list_children should not attempt to refresh image (issue#21670, pr#18114, Jason Dillaman)
  • librbd: minor cleanup of the IO pathway (pr#20560, Jason Dillaman)
  • librbd: minor code cleanup (pr#21165, songweibin)
  • librbd: missing ‘return’ in deep_copy::ObjectCopyRequest::send_read_object (pr#21493, Mykola Golub)
  • librbd: new tag should use on-disk committed position (issue#22945, pr#20423, Jason Dillaman)
  • librbd: object map batch update might cause OSD suicide timeout (issue#21797, pr#18315, Jason Dillaman)
  • librbd: possible deadlock with synchronous maintenance operations (issue#22120, pr#18909, Jason Dillaman)
  • librbd: potential crash if object map check encounters error (issue#22819, pr#20214, Jason Dillaman)
  • librbd: potential race between discard and writeback (pr#21248, Jason Dillaman)
  • librbd: potential race in RewatchRequest when resetting watch_handle (pr#20420, Mykola Golub)
  • librbd: prefer templates to macros (pr#19912, Adam C. Emerson)
  • librbd: prevent overflow of discard API result code (issue#21966, pr#18923, Jason Dillaman)
  • librbd: prevent watcher from unregistering with in-flight actions (issue#23955, pr#21763, Jason Dillaman)
  • librbd: refresh image after applying new metadata (issue#21711, pr#18158, Jason Dillaman)
  • librbd: release lock executing deep copy progress callback (issue#23929, pr#21727, Mykola Golub)
  • librbd: remove unused member in FlattenRequest (pr#19416, Mykola Golub)
  • librbd: remove unused variables from ReadResult refactor (pr#18277, Jason Dillaman)
  • librbd: rename of non-existent image results in seg fault (issue#21248, pr#17502, Jason Dillaman)
  • librbd: set deleted parent pointer to null (issue#22158, pr#19003, Jason Dillaman)
  • librbd: should not set self as remote peer (pr#17300, songweibin)
  • librbd: small cleanup for recently merged code (pr#20578, Mykola Golub)
  • librbd: snapshots should be created/removed against data pool (issue#21567, pr#18043, Jason Dillaman)
  • librbd: speed up object map disk usage and resize (pr#20218, shun-s)
  • librbd: speed up sparse copy when object map is available (pr#18967, Song Shun)
  • librbd: update mirror::EnableRequest diagram according to code (pr#19130, Mykola Golub)
  • librbd: use steady clock to measure elapsed time in AioCompletion (pr#20007, Mohamad Gebai)
  • librbd: validate if dst group snap name is the same with src (pr#20395, songweibin)
  • log: Fix AddressSanitizer: new-delete-type-mismatch (issue#23324, pr#20930, Brad Hubbard)
  • log: fix build on osx (pr#18213, Kefu Chai)
  • log: silence warning from -Wsign-compare (pr#18326, Jos Collin)
  • log: Use the coarse real time clock in log timestamps (pr#18141, Adam C. Emerson)
  • mds: check metadata pool not cluster is full (issue#22483, pr#19602, Patrick Donnelly)
  • mds: fix CEPH_STAT_RSTAT definition (pr#21633, “Yan, Zheng”)
  • mds: get rid of the “if” check which is unnecessary inside a loop (pr#18904, dongdong tao)
  • mds: Remove redundant null pointer check (pr#19750, Brad Hubbard)
  • mds: simplify the code logic in replay_alloc_ids (pr#18893, dongdong tao)
  • mempool: fix lack of pool names in mempool:dump output for JSON format (pr#18329, Igor Fedotov)
  • messages: Initialization of uninitialized members various classes (pr#16848, amitkuma)
  • messages/MDentryLink: add const to member function (pr#15479, yonghengdexin735)
  • messages,test,msg: initialize h,reply_type,owner (pr#17767, Amit Kumar)
  • mgr: add mgr daemon to DaemonStateIndex with metadata (hostname) (issue#23286, pr#20875, Jan Fajerski)
  • mgr: add missing call to pick_addresses (issue#20955, pr#16940, John Spray)
  • mgr: add the ip addr of standbys (pr#16476, huanwen ren)
  • mgr: add units to performance counters (issue#22747, pr#20152, Rubab Syed)
  • mgr: allow service daemons to unregister from ServiceMap (pr#20761, Sage Weil)
  • mgr: apply a threshold to perf counter prios (pr#16699, John Spray)
  • mgr: balancer: fixed mistype “AttributeError: ‘Logger’ object has no attribute ‘err’” (pr#20130, Konstantin Shalygin)
  • mgr: centralized setting/getting of mgr configs (pr#21442, John Spray, Rubab Syed)
  • mgr: ceph-mgr: can not change prometheus port for mgr (pr#17746, wujian)
  • mgr: common interface for TSDB modules (pr#17735, Jan Fajerski, John Spray, My Do)
  • mgr/dashboard: Adapt help text if server_addr is not set (pr#21640, Volker Theile)
  • mgr/dashboard: Adapt RBD form to new application_metadata type (pr#21602, Volker Theile)
  • mgr/dashboard: Add Api module (pr#21126, Tiago Melo)
  • mgr/dashboard: Add ‘autofocus’ directive (pr#21559, Volker Theile)
  • mgr/dashboard: Add CdDatePipe (pr#21087, Ricardo Marques)
  • mgr/dashboard: Add ‘cd-error-panel’ component to display error messages (pr#21558, Volker Theile)
  • mgr/dashboard: Add ‘cd-loading-panel’ component (pr#21618, Volker Theile)
  • mgr/dashboard: Add custom validators (pr#21041, Volker Theile)
  • mgr/dashboard: Add DimlessBinaryDirective (pr#20972, Ricardo Marques)
  • mgr/dashboard: Add ErasureCodeProfile controller (issue#23345, pr#20920, Sebastian Wagner, Stephan Müller)
  • mgr/dashboard: Add ‘forceIdentifier’ attribute to datatable (pr#21497, Volker Theile)
  • mgr/dashboard: Add helper component (pr#20971, Ricardo Marques)
  • mgr/dashboard: additional fixes to block pages (pr#20941, Jason Dillaman)
  • mgr/dashboard: Add minimalistic browsable API (pr#20873, Sebastian Wagner)
  • mgr/dashboard: Add notification service/component (pr#21078, Tiago Melo)
  • mgr/dashboard: Add Pool-create to the backend (issue#23345, pr#20865, Sebastian Wagner)
  • mgr/dashboard: Add RGW user and bucket management features (pr#21351, Volker Theile)
  • mgr/dashboard: Adds reusable deletion dialog (pr#20899, Stephan Müller, Tiago Melo)
  • mgr/dashboard: Add submit button component (pr#21011, Tiago Melo)
  • mgr/dashboard: Add usage bar component (pr#21128, Ricardo Marques)
  • mgr/dashboard: Angular modules cleanup (pr#21402, Tiago Melo)
  • mgr/dashboard: Asynchronous tasks (frontend) (pr#20962, Ricardo Marques)
  • mgr/dashboard: awsauth: fix python3 string decode problem (pr#21875, Ricardo Dias)
  • mgr/dashboard: Change font-family of checkbox (pr#21787, Tiago Melo)
  • mgr/dashboard: Clean up Pylint warnings (pr#21694, Sebastian Wagner)
  • mgr/dashboard: Convert floating values to bytes (pr#21677, Stephan Müller)
  • mgr/dashboard: Convert the RBD feature names to a list of strings (pr#21024, Tatjana Dehler)
  • mgr/dashboard: Deletion dialog falsely executes deletion when pressing ‘Cancel’ (pr#22032, Volker Theile)
  • mgr/dashboard: Display notification if RGW is not configured (pr#21977, Volker Theile)
  • mgr/dashboard: Display RBD form errors on submission (pr#21529, Ricardo Marques)
  • mgr/dashboard: Enable object rendering in KV-table (pr#21701, Stephan Müller)
  • mgr/dashboard: fix 500 error on block device iSCSI status page (pr#20928, Jason Dillaman)
  • mgr/dashboard: fix dashboard python 3 support (pr#21007, Ricardo Dias)
  • mgr/dashboard: Fix data race and use-before-assignment (pr#21590, Sebastian Wagner)
  • mgr/dashboard: fixed password generation in Auth controller (issue#23404, pr#21006, Ricardo Dias)
  • mgr/dashboard: Fixes documentation link- to open in new tab (pr#22262, Kanika Murarka)
  • mgr/dashboard: Fixes type error in RBD form (pr#21681, Stephan Müller)
  • mgr/dashboard: fix frontend e2e tests (pr#20943, Tiago Melo)
  • mgr/dashboard: fix FS status on old MDS daemons (issue#20692, pr#16960, John Spray)
  • mgr/dashboard: fix linting problem (pr#22277, Tiago Melo)
  • mgr/dashboard: Fix missing $event on deletion modal (pr#21667, Ricardo Marques)
  • mgr/dashboard: Fix moment.js deprecation warning (pr#22052, Tiago Melo)
  • mgr/dashboard: Fix objects named default are inaccessible (pr#20976, Sebastian Wagner)
  • mgr/dashboard: Fix RBD task metadata (pr#22152, Tiago Melo)
  • mgr/dashboard: Fix table without fetchData (pr#21086, Ricardo Marques)
  • mgr/dashboard: Fix the data table action selector (pr#21270, Stephan Müller)
  • mgr/dashboard: fix two type errors found by mypy (pr#21774, Sebastian Wagner)
  • mgr/dashboard: Handle errors during deletion (pr#22029, Volker Theile)
  • mgr/dashboard: Implement a RGW proxy (pr#21258, Volker Theile, Patrick Nawracay)
  • mgr/dashboard: Improve background tasks style (pr#21462, Ricardo Marques)
  • mgr/dashboard: improve error handling (pr#18182, Nick Erdmann)
  • mgr/dashboard: Improve error panel (pr#21978, Volker Theile)
  • mgr/dashboard: Improve npm start script (pr#20989, Ricardo Marques)
  • mgr/dashboard: Improve table search (pr#20807, Stephan Müller)
  • mgr/dashboard: Load the datatable content on component initialization (pr#21595, Volker Theile)
  • mgr/dashboard: Navbar dropdown button does not respond for mobile browsers (pr#21979, Volker Theile)
  • mgr/dashboard: Notification improvements (pr#21350, Tiago Melo)
  • mgr/dashboard: pool: fix python3 dict_keys error (pr#21636, Ricardo Dias)
  • mgr/dashboard: Pool listing (pr#21353, Stephan Müller)
  • mgr/dashboard: rbd: add @AuthRequired to snapshots controller (pr#21517, Ricardo Dias)
  • mgr/dashboard: RBD copy, RBD flatten and snapshot clone (frontend) (pr#21526, Ricardo Marques, Ricardo Dias)
  • mgr/dashboard: RBD management (frontend) (pr#21385, Ricardo Marques)
  • mgr/dashboard: Refactor multiple duplicates of get_rate() (pr#21022, Sebastian Wagner)
  • mgr/dashboard: Refactor RGW backend (pr#21855, Volker Theile)
  • mgr/dashboard: Rename and refactor ApiInterceptorService class (pr#21386, Volker Theile)
  • mgr/dashboard: Replace font-awesome with fork-awesome (pr#21327, Lenz Grimmer)
  • mgr/dashboard: restcontroller: fix detection of id args in element requests (pr#21290, Ricardo Dias)
  • mgr/dashboard: RESTController improvements (pr#21516, Ricardo Dias)
  • mgr/dashboard: run-tox: pass CEPH_BUILD_DIR value into tox script (pr#21445, Ricardo Dias)
  • mgr: dashboard: show per pool IOPS on health page (#22495) (issue#22495, pr#19981, Konstantin Shalygin)
  • mgr/dashboard: Support aditional info on ‘cd-view-cache’ (pr#21060, Ricardo Marques)
  • mgr/dashboard: TaskManager bug fixes (pr#21240, Ricardo Dias)
  • mgr/dashboard: Update selected items on table refresh (pr#21099, Ricardo Marques)
  • mgr/dashboard: Use Bootstrap CSS (pr#21780, Volker Theile)
  • mgr/dashboard: using RoutesDispatcher as HTTP request dispatcher (pr#21239, Ricardo Dias)
  • mgr/dashboard_v2: add mgr to the list of perf counters (pr#20783, Tiago Melo)
  • mgr/dashboard_v2: add mocked service provider for TcmuIscsiService (pr#20775, Tiago Melo)
  • mgr/dashboard_v2: Add toggle able columns (pr#20806, Stephan Müller)
  • mgr/dashboard_v2: Configuration settings support (pr#20743, Ricardo Dias)
  • mgr/dashboard_v2: fix and improve table details (pr#20811, Tiago Melo)
  • mgr/dashboard_v2: Fix cephfs template table usage (pr#20804, Stephan Müller)
  • mgr/dashboard_v2: fix cluster configuration page (pr#20821, Tiago Melo)
  • mgr/dashboard_v2: Improve charts tooltips (pr#20757, Tiago Melo)
  • mgr/dashboard_v2: Pool controller (pr#20823, Ricardo Dias)
  • mgr/dashboard_v2: Rotate the refresh icon on load (pr#20805, Stephan Müller)
  • mgr: die on bind() failure (pr#20595, John Spray)
  • mgr: disconnect unregistered service daemon when report received (issue#22286, pr#19261, Jason Dillaman)
  • mgr: emit cluster log message on serve() exception (issue#21999, pr#18672, John Spray)
  • mgr: Expose rgw perf counters (pr#21269, Boris Ranto)
  • mgr: fix “access denied” message (pr#19518, John Spray)
  • mgr: fix crashable DaemonStateIndex::get calls (issue#17737, pr#17933, John Spray)
  • mgr: fix crash in MonCommandCompletion (issue#21157, pr#17308, John Spray)
  • mgr: fixes python error handling (issue#23406, pr#21005, Ricardo Dias)
  • mgr: fix MSG_MGR_MAP handling (pr#20892, Gu Zhongyan)
  • mgr: fix “osd status” command exception if OSD not in pgmap stats (issue#21707, pr#18173, Yanhu Cao)
  • mgr: fix py3 support (issue#22880, pr#20362, Kefu Chai)
  • mgr: fix py calls for dne service perf counters (issue#21253, pr#17605, John Spray)
  • mgr: implement completion of osd MetadataUpdate (issue#21159, pr#16925, Yanhu Cao)
  • mgr: implement ‘osd safe-to-destroy’ and ‘osd ok-to-stop’ commands (pr#16976, Sage Weil)
  • mgr: improved module loading for error reporting etc (issue#21999, issue#21683, issue#21502, pr#19235, John Spray)
  • mgr: improve reporting on unloadable modules (issue#23358, pr#20921, John Spray)
  • mgr: increase time resolution of Commit/Apply OSD latencies (pr#19232, Коренберг Марк)
  • mgr: initialize PyModuleRegistry sooner (issue#22918, pr#20321, John Spray)
  • mgr: In plugins ‘module’ classes need not to be called “Module” anymore (issue#17454, pr#18526, Kefu Chai, bhavishyagopesh)
  • mgr: locking fixes (issue#21158, pr#17309, John Spray)
  • mgr: mgr/balancer: cast config vals to int or float (issue#22429, pr#19493, Dan van der Ster)
  • mgr: mgr/balancer: don’t use ‘foo’ tags on commands (issue#22361, pr#19482, John Spray)
  • mgr: mgr/balancer: fix KeyError in balancer rm (issue#22470, pr#19578, Dan van der Ster)
  • mgr: mgr/balancer: fix OPTIONS definition (pr#21620, John Spray)
  • mgr: mgr/balancer: fix upmap; default balancer module enabled (pr#18691, Sage Weil)
  • mgr: mgr/balancer: make crush-compat mode work (pr#17983, Sage Weil)
  • mgr: mgr/balancer: mgr module to automatically balance PGs across OSDs (pr#16272, Spandan Kumar Sahu, Sage Weil)
  • mgr: mgr/balancer: more pool-specific enhancements (pr#20225, xie xingguo)
  • mgr: mgr/balancer: pool-specific optimization support and bug fixes (pr#20154, xie xingguo)
  • mgr: mgr/balancer: replace magic value of -1 for DEFAULT_CHOOSE_ARGS (pr#20258, Kefu Chai)
  • mgr: mgr/balancer: skip CRUSH_ITEM_NONE (pr#18894, Sage Weil)
  • mgr: mgr/balancer: two more fixes (pr#20180, xie xingguo)
  • mgr: mgrc: free MMgrClose in handle_mgr_close (issue#23846, pr#21626, Casey Bodley)
  • mgr: mgr/DaemonServer: add overrides value to ‘config show’ (pr#21093, Gu Zhongyan)
  • mgr: mgr/DaemonServer.cc: [Cleanup] Change to using get_val template function (pr#18717, Shinobu Kinjo)
  • mgr: mgr/DaemonServer: [Cleanup] Remove redundant code (pr#18716, Shinobu Kinjo)
  • mgr: mgr/dashboard: add configuration setting browser (issue#22522, pr#20043, Rubab Syed)
  • mgr: mgr/dashboard: add image id to mgr rbd info instead of block_name_prefix (pr#20884, zouaiguo)
  • mgr: mgr/dashboard: Add monitor list (pr#19632, Rubab Syed)
  • mgr: mgr/dashboard: Add RGW user and bucket lists (read-only) (pr#20869, Volker Theile)
  • mgr: mgr/dashboard: add TLS (pr#21627, John Spray)
  • mgr: mgr/dashboard: Add toBytes() method to FormatterService (pr#20978, Volker Theile)
  • mgr: mgr/dashboard: asynchronous task support (pr#20870, Ricardo Dias)
  • mgr: mgr/dashboard: change raw usage chart’s color depending on usage (pr#17421, Nick Erdmann)
  • mgr: mgr/dashboard: fix audit log loading (pr#18848, John Spray)
  • mgr: mgr/dashboard: Fix backend tests for newer CherryPy versions (pr#20778, Patrick Nawracay)
  • mgr: mgr/dashboard: Fix PG status coloring (pr#19431, Wido den Hollander)
  • mgr: mgr/dashboard: format tooltip’s label as user friendly string (pr#18769, Yao Zongyou)
  • mgr: mgr/dashboard: handle null in format_number (issue#21570, pr#17991, John Spray)
  • mgr: mgr/dashboard: HTTP request logging (pr#20797, Ricardo Dias)
  • mgr: mgr/dashboard: Improve auth interceptor (pr#20847, Volker Theile)
  • mgr: mgr/dashboard: performance counter browsers (issue#22521, pr#19922, Rubab-Syed)
  • mgr: mgr/dashboard: RBD management (backend) (pr#21360, Ricardo Dias)
  • mgr: mgr/dashboard: Remove unused code (pr#21045, Volker Theile)
  • mgr: mgr/dashboard: Remove useless code (pr#20958, Volker Theile)
  • mgr: mgr/dashboard: show warnings if data is out of date or mons are down (pr#18847, John Spray)
  • mgr: mgr/dashboard: sort servers and OSDs in OSD list (issue#21572, pr#17993, John Spray)
  • mgr: mgr/dashboard: use rel=”icon” for favicon (pr#18013, Kefu Chai)
  • mgr: mgr/dashboard v2: Add CSS class for required form fields (pr#20747, Volker Theile)
  • mgr: mgr/dashboard_v2: Add RBD create functionality to the Python backend (pr#20751, Tatjana Dehler)
  • mgr: mgr/dashboard v2: Add units to performance counters (pr#20742, Volker Theile)
  • mgr: mgr/dashboard v2: Display loading indicator in datatables during first load (pr#20744, Volker Theile)
  • mgr: mgr/dashboard v2: Don’t show details if multiple OSDs are selected (pr#20772, Volker Theile)
  • mgr: mgr/dashboard v2: implement can_run method (pr#20728, John Spray)
  • mgr: mgr/dashboard_v2: Initial submission of a web-based management UI (replacement for the existing dashboard) (pr#20103, Stephan Müller, Lenz Grimmer, Tiago Melo, Ricardo Marques, Sebastian Wagner, Patrick Nawracay, Ricardo Dias, Volker Theile, Kai Wagner, Tatjana Dehler)
  • mgr: mgr/dashboard v2: Introduce CdTableSelection model (pr#20746, Volker Theile)
  • mgr: mgr/dashboard_v2: Removed unused tools.detail_route() (pr#20765, Sebastian Wagner)
  • mgr: mgr/influx: Added Additional Stats (pr#21424, mhdo2)
  • mgr: mgr/influx: Add InfluxDB SSL Option (pr#19374, Tobias Gall)
  • mgr: mgr/influx: Only split string on first occurence of dot (.) (issue#23996, pr#21795, Wido den Hollander)
  • mgr: mgr/influx: PEP-8 and other fixes to Influx module (pr#19229, Wido den Hollander)
  • mgr: mgr/influx: Various fixes and improvements (pr#20187, Wido den Hollander)
  • mgr: mgr/influx: Various time fixes (pr#20494, Wido den Hollander)
  • mgr: mgr/localpool: default to 3x; allow min_size adjustment (pr#18089, Sage Weil)
  • mgr: mgr/MgrClient: guard send_pgstats() with lock (issue#23370, pr#20909, Kefu Chai)
  • mgr: mgr/MgrClient: service registration filtered by service name instead of daemon name (pr#21459, runsisi)
  • mgr: mgr/PGMap: drop REQUEST_{SLOW,STUCK} HEALTH_WARNs (pr#19114, Kefu Chai)
  • mgr: mgr/prometheus: add ceph_disk_occupation series (issue#21594, pr#18021, John Spray)
  • mgr: mgr/prometheus: add missing ‘deep’ state to PG_STATES in ceph-mgr prometheus plugin (issue#22116, pr#18890, Peter Woodman)
  • mgr: mgr/prometheus: Fix for MDS metrics (issue#20899, pr#17318, John Spray, Jeremy H Austin)
  • mgr: mgr/prometheus: fix PG state names (pr#21288, John Spray)
  • mgr: mgr/prometheus: Skip bogus entries (pr#20456, Boris Ranto)
  • mgr: mgr/prometheus: skip OSD output if missing from CRUSH devices (pr#20644, John Spray)
  • mgr: mgr/restful: A couple of restful fixes (pr#18649, Boris Ranto)
  • mgr: mgr/restful: cleaner message when not configured (issue#21292, pr#17573, John Spray)
  • mgr: mgr/smart: fix python3 module loading (pr#21047, Ricardo Dias)
  • mgr: mgr/status: fix ceph fs status returns error (issue#21752, pr#18233, Yanhu Cao)
  • mgr: mgr/status: format byte quantities in base 2 multiples (issue#21189, pr#17380, John Spray)
  • mgr: mgr/telemetry: Add Ceph Telemetry module to send reports back to project (pr#21970, Wido den Hollander)
  • mgr: mgr/zabbix: fix div by zero (issue#21518, pr#17931, John Spray)
  • mgr: mgr/zabbix: ignore osd with 0 kb capacity (issue#21904, pr#18809, Ilja Slepnev)
  • mgr: mgr/zabbix: Implement health checks (pr#20198, Wido den Hollander)
  • mgr: mgr/zabbix: Send max, min and avg PGs of OSDs to Zabbix (pr#21043, Wido den Hollander)
  • mgr: mgr/Zabbix: Various fixes to Zabbix module (pr#19452, Wido den Hollander)
  • mgr: mimic: mgr/telegraf: Telegraf module for Ceph Mgr (pr#22013, Wido den Hollander)
  • mgr: Modify mgr-influx module database check to not require admin privileges (pr#18102, Benjeman Meekhof)
  • mgr: mon,mgr: improve ‘mgr module disable’ cmd (pr#21188, Gu Zhongyan)
  • mgr: mon, mgr: move “osd pool stats” command to mgr and mgr python module (pr#19985, Chang Liu)
  • mgr: mon/MgrStatMonitor: fix formatting of pending_digest (issue#22991, pr#20426, Patrick Donnelly)
  • mgr,mon: mon/MgrMonitor: read cmd descs if empty on update_from_paxos() (issue#21300, pr#17846, Joao Eduardo Luis)
  • mgr,mon: mon,mgr: remove single wildcard ‘*’ from ceph comand line description (pr#21139, Gu Zhongyan)
  • mgr,mon: mon/mgr: sync “mgr_command_descs”,”osd_metadata” and “mgr_metadata” prefixes to new mons (issue#21527, pr#17929, huanwen ren)
  • mgr,mon: mon/MonCommands: mgr metadata - improve parameter naming consistency (issue#23330, pr#20866, Jan Fajerski)
  • mgr: preventing blank hostname in DaemonState (issue#20887, issue#21060, pr#17138, liuchang0812)
  • mgr: prometheus: added osd commit/apply latency metrics (#22718) (issue#22718, pr#19980, Konstantin Shalygin)
  • mgr: prometheus: Don’t crash on OSDs without metadata (pr#20539, Christopher Blum)
  • mgr: prometheus fix metadata labels (pr#21557, Jan Fajerski)
  • mgr: prometheus: set metadata metrics value to ‘1’ (#22717) (issue#22717, pr#19979, Konstantin Shalygin)
  • mgr: pybind/mgr/balancer: add sanity check against empty adjusted_map (pr#20836, xie xingguo)
  • mgr: pybind/mgr/balancer: fix pool-deletion vs auto-optimization race (pr#20706, xie xingguo)
  • mgr: pybind/mgr/balancer: fix sanity check against empty weight-set (pr#20278, xie xingguo)
  • mgr: pybind/mgr/balancer: increase bad_steps properly (pr#20194, xie xingguo)
  • mgr: pybind/mgr/balancer: load weight-set from ms (pr#20197, xie xingguo)
  • mgr: pybind/mgr/balancer: more specific command outputs (pr#20305, xie xingguo)
  • mgr: pybind/mgr/balancer: remove optimization plan properly (pr#20224, xie xingguo)
  • mgr: pybind/mgr/balancer: two more fixes (pr#20788, xie xingguo)
  • mgr: pybind/mgr/dashboard: add url_prefix (issue#20568, pr#17119, Nick Erdmann)
  • mgr: pybind/mgr/dashboard: fix duplicated slash in html href (issue#22851, pr#20229, Shengjing Zhu)
  • mgr,pybind: mgr/dashboard: fix pool size base conversion (pr#16771, Yixing Yan)
  • mgr: pybind/mgr/dashboard: fix reverse proxy support (issue#22557, pr#19758, Nick Erdmann)
  • mgr,pybind: mgr/iostat: print output as a table (pr#21338, Mohamad Gebai)
  • mgr: pybind/mgr/localpool: module to automagically create localized pools (pr#17528, Sage Weil)
  • mgr: pybind/mgr/mgr_module: add default param for MgrStandbyModule.get_con… (pr#19948, Kefu Chai)
  • mgr: pybind/mgr/mgr_module: make rados handle available to all modules (pr#19972, Sage Weil)
  • mgr: pybind/mgr_module: move PRIO_* and PERFCOUNTER_* to MgrModule class (pr#18251, Jan Fajerski)
  • mgr: pybind/mgr: new ‘hello world’ mgr module skeleton (pr#19491, Yaarit Hatuka)
  • mgr: pybind/mgr/prometheus: add file_sd_config command (pr#21061, Jan Fajerski)
  • mgr: pybind/mgr/prometheus: add osd_in/out metric; make osd_weight a metric (pr#18243, Jan Fajerski)
  • mgr: pybind/mgr/prometheus: add StandbyModule and handle failed MON cluster (pr#19744, Jan Fajerski)
  • mgr: pybind/mgr/prometheus: don’t crash when encountering an unknown PG state (pr#18903, Jan Fajerski)
  • mgr: pybind/mgr/prometheus: don’t export metrics for dead daemon; new metrics (pr#20506, Jan Fajerski)
  • mgr: pybind/mgr/prometheus: fix creation of osd_metadata metric (pr#21530, Jan Fajerski)
  • mgr: pybind/mgr/prometheus: fix metric type undef -> untyped (issue#22313, pr#19524, Ilya Margolin)
  • mgr: pybind/mgr/prometheus: fix metric type undef -> untyped (pr#18208, Jan Fajerski)
  • mgr,pybind: pybing/mgr/prometheus: return default port if config-key get returns … (pr#21696, Jan Fajerski)
  • mgr: python interface rework + enable modules to run in standby mode (issue#21593, issue#17460, pr#16651, John Spray, Sage Weil)
  • mgr: quieten logging on missing OSD stats (pr#20485, John Spray)
  • mgr,rbd: mgr/dashboard: added iSCSI IOPS/throughput metrics (issue#21391, pr#18653, Jason Dillaman)
  • mgr,rbd: mgr/dashboard: fix duplicate images listed on iSCSI status page (issue#21017, pr#17055, Jason Dillaman)
  • mgr: reconcile can_run checks and selftest (pr#21607, John Spray, Kefu Chai)
  • mgr: remove a few junk lines (pr#20005, John Spray)
  • mgr: remove unused static files from dashboard module (pr#16762, John Spray)
  • mgr: request daemon’s metadata when receiving a report from an unknown server (issue#21687, pr#18484, Chang Liu)
  • mgr,rgw: mgr/dashboard: RGW page (pr#19512, Chang Liu)
  • mgr,rgw: prometheus: Implement rgw_metadata metric (pr#21383, Boris Ranto)
  • mgr: safety checks on pyThreadState usage (pr#18093, John Spray)
  • mgr: set explicit thread name (issue#21404, pr#17756, John Spray)
  • mgr: silence warning from -Wsign-compare (pr#17881, Jos Collin)
  • mgr: skip first non-zero incremental in PGMap::apply_incremental() (issue#21773, pr#18347, Aleksei Gutikov)
  • mgr/status: output to stdout, not stderr (issue#24175, pr#22135, John Spray)
  • mgr: store declared_types in MgrSession (issue#21197, pr#17932, John Spray)
  • mgr: systemd: Wait 10 seconds before restarting ceph-mgr (issue#23083, pr#20533, Wido den Hollander)
  • mgr,tests: mgr/dashboard: skip data pool testcase for none-bluestore clusters (pr#21004, Tatjana Dehler)
  • mgr,tests: mgr/dashboard_v2: fix test_perf_counters_mgr_get (pr#20916, Tiago Melo)
  • mgr,tests: qa: add new prometheus test to rados/mgr suite (pr#20047, John Spray)
  • mgr,tests: qa: configure zabbix properly before selftest (issue#22514, pr#19634, John Spray)
  • mgr,tests: qa: fix mgr _load_module helper (pr#18685, John Spray)
  • mgr,tools: mgr/iostat: implement ‘ceph iostat’ as a mgr plugin (pr#20100, Mohamad Gebai)
  • mgr: use new style config opts + add metadata (pr#17374, John Spray)
  • mgr/zabbix: Fix wrong log message (pr#21237, Gu Zhongyan)
  • mgr/zabbix: monitoring template improvements (pr#19901, Marc Schoechlin)
  • mon: Add ceph osd get-require-min-compat-client command (pr#19015, hansbogert)
  • mon: add ‘ceph osd pool get erasure allow_ec_overwrites’ command (pr#21102, Mykola Golub)
  • mon: add MMonHealth back (issue#22462, pr#20528, Kefu Chai)
  • mon: add mon_health_preluminous_compat_warning (pr#16902, Sage Weil)
  • mon: a few conversions to monotonic clock (pr#18595, Patrick Donnelly)
  • mon: align lspools output (pr#19597, Jos Collin)
  • mon: allow cluter and debug logs to go to stderr, with appropriate prefix (pr#19385, Sage Weil)
  • mon: cache reencoded osdmaps (issue#23713, pr#21605, Sage Weil, Xiaoxi CHEN)
  • mon: centralized config (pr#20172, Sage Weil)
  • mon: “ceph osd crush rule rename” support (pr#17029, xie xingguo)
  • mon: check monitor address configuration (pr#18073, Li Wang)
  • mon: clean up cluster logging on mon events (issue#22082, pr#18822, John Spray)
  • mon: cleanups to optracker code (pr#21371, John Spray)
  • mon: cleanup unused option mon_health_data_update_interval (pr#17728, Yao Guotao)
  • mon: common/options: set max_background_jobs instead of max_background_compactions (pr#18397, Kefu Chai)
  • mon: Compress the warnings of pgs not scrubbed or deep-scrubbed (pr#17295, Zhi Zhang)
  • mon: do not use per_pool_sum_delta to show recovery summary (issue#22727, pr#20009, Chang Liu)
  • mon: don’t blow away bootstrap-mgr on upgrades (issue#20950, pr#18399, John Spray)
  • mon: double mon_mgr_mkfs_grace from 60s -> 120s (pr#20955, Sage Weil)
  • mon: Drop redundant access specifier, etc (cleanup) (pr#19028, Shinobu Kinjo)
  • mon: dump percent_used PGMap field as float (pr#20439, John Spray)
  • mon: dump servicemap along with MgrStatMonitor dump info (pr#18760, Zhi Zhang)
  • mon: expand cap validity check for mgr, osd, mds (issue#22525, pr#21311, Jing Li, Sage Weil)
  • mon: final luminous compatset feature and osdmap flag (pr#17333, Sage Weil)
  • mon: fix commands advertised during mon cluster upgrade (pr#16871, Sage Weil)
  • mon: fix dropping mgr metadata for active mgr (#21260) (issue#21260, pr#17571, John Spray)
  • mon: fix “fs new” pool metadata update, tests (issue#20959, pr#16954, Greg Farnum)
  • mon: fix legacy health checks in ‘ceph status’ during upgrade; fix jewel-x upgrade combo (pr#16967, Sage Weil)
  • mon: fix mgr using auth_client_required policy (pr#20048, John Spray)
  • mon: fix osd out clog message (issue#21249, pr#17525, John Spray)
  • mon: fix slow op warning on mon, improve slow op warnings (issue#23769, pr#21684, Sage Weil)
  • mon: fix structure of ‘features’ command (pr#20115, Sage Weil)
  • mon: fix two stray legacy get_health() callers (pr#17269, Sage Weil)
  • mon: fix wrong mon-num counting logic of ‘ceph features’ command (pr#16887, xie xingguo)
  • mon: handle bad snapshot removal reqs gracefully (issue#18746, pr#20835, Paul Emmerich)
  • mon: handle monitor lag when killing mgrs (issue#20629, pr#18268, John Spray)
  • mon: incorrect MAX AVAIL in “ceph df” (issue#21243, pr#17513, liuchang0812)
  • mon: invalid JSON returned when querying pool parameters (issue#23200, pr#20745, Chang Liu)
  • mon/LogMonitor: call no_reply() on ignored log message (pr#22104, Sage Weil)
  • mon: mark mgr reports as no_reply (issue#22114, pr#21057, Kefu Chai)
  • mon: mark mon_allow_pool_delete as observed (pr#18125, Dan van der Ster)
  • mon: mark OSD beacons and pg_create messages as no_reply (issue#22114, pr#20517, Greg Farnum)
  • mon: mon/AuthMonitor: don’t validate fs authorize caps with valid_caps() (pr#21418, Joao Eduardo Luis)
  • mon: mon/ConfigMonitor: clean up prepare_command() (pr#20911, Gu Zhongyan)
  • mon: mon/Elector: force election epoch bump on start (issue#20949, pr#16944, Sage Weil)
  • mon: mon/Elector: remove unused member fields start_stamp and ack_stamp (pr#21091, runsisi)
  • mon: mon/LogMonitor: “log last” should return up to n entries (pr#18759, Kefu Chai)
  • mon: mon/MDSMonitor: fix clang build failure (pr#20637, Willem Jan Withagen)
  • mon: mon,mgr: make osd_metric more popular and report slow ops to mgr (issue#23045, pr#20660, lvshanchun)
  • mon: mon/MgrMonitor: limit mgrmap history (issue#22257, pr#19185, Sage Weil)
  • mon: mon/MonCommands: fix copy-and-paste error (pr#17271, xie xingguo)
  • mon: mon,option: set default value for mon_dns_srv_name (issue#21204, pr#17539, Kefu Chai)
  • mon: mon/OSDMonitor: add location option for “crush add-bucket” command (pr#17125, xie xingguo)
  • mon: mon/OSDMonitor: add ‘osd crush set-all-straw-buckets-to-straw2’ (pr#18460, Sage Weil)
  • mon: mon/OSDMonitor: add plain output for “crush class ls-osd” command (pr#17034, xie xingguo)
  • mon: mon/OSDMonitor: add space after __func__ in log msg (pr#19127, Kefu Chai)
  • mon: mon/OSDMonitor: Better prepare_command_pool_set E2BIG error message (pr#19944, Brad Hubbard)
  • mon: mon/OSDMonitor.cc: fix expected_num_objects interpret error (issue#22530, pr#19651, Yang Honggang)
  • mon: mon/OSDMonitor.cc : set erasure-code-profile to “” when create replicated pools (pr#19673, zouaiguo)
  • mon: mon/OSDMonitor: check last_scan_epoch instead when sending creates (issue#20785, pr#17248, Kefu Chai)
  • mon: mon/OSDMonitor: clean up cmd ‘osd tree-from’ (pr#20839, Gu Zhongyan)
  • mon: mon/OSDMonitor: do not send_pg_creates with stale info (issue#20785, pr#17065, Kefu Chai)
  • mon: mon/OSDMonitor: error out if setting ruleset-* ec profile property (pr#17848, Sage Weil)
  • mon: mon/OSDMonitor: fix improper input/testing range of crush somke testing (pr#17179, xie xingguo)
  • mon: mon/OSDMonitor: fix ‘osd pg temp’ unable to cleanup pg-temp (pr#16892, xie xingguo)
  • mon: mon/OSDMonitor: implement ‘osd crush ls <node>’ (pr#16920, Sage Weil)
  • mon: mon/OSDMonitor: kill pending upmap changes too if pool is gone (pr#20704, xie xingguo)
  • mon: mon/OSDMonitor: logging non-active osd id when handling osd beacon (pr#21092, runsisi)
  • mon: mon/OSDMonitor: make ‘osd crush rule rename’ idempotent (issue#21162, pr#17329, xie xingguo)
  • mon: mon/OSDMonitor: “osd pool application get” support (issue#20976, pr#16955, xie xingguo)
  • mon: mon/OSDMonitor: txsize should be greater or eq to prune_interval - 1 (pr#21430, Kefu Chai)
  • mon: mon/PGMap: drop DISK LOG column (pr#17617, Sage Weil)
  • mon: mon/PGMap: fix “0 stuck requests are blocked > 4096 sec” warn (pr#17099, xie xingguo)
  • mon: mon/PGMap: nice numbers for ‘data’ section of ‘ceph df’ command (pr#17368, xie xingguo)
  • mon: mon/PGMap: Remove unnecessary header (pr#18343, Shinobu Kinjo)
  • mon: mon/PGMap: reweight::by_utilization - skip DNE osds (issue#20970, pr#17064, xie xingguo)
  • mon: mon/pgmap: update pool nearfull display (pr#17043, huanwen ren)
  • mon: more aggressively convert crush rulesets -> distinct rules (pr#17508, John Spray, Sage Weil)
  • mon: more constness (pr#17748, Kefu Chai)
  • mon: node ls improvement (pr#20820, Gu Zhongyan)
  • mon: ‘node ls’ mgr support (pr#20711, Gu Zhongyan)
  • mon: NULL check of logger before use (pr#18788, Amit Kumar)
  • mon,osd: dump “compression_algorithms” in “mon metadata” (issue#24135, issue#22420, pr#22004, Kefu Chai, Casey Bodley)
  • mon: osd feature checks with 0 up osds (issue#21471, issue#20751, pr#17831, Brad Hubbard, Sage Weil)
  • mon: osdmap prune (pr#19331, Joao Eduardo Luis)
  • mon/OSDMonitor: cleanup: move bufferlist before use (pr#18258, Shinobu Kinjo)
  • mon/OSDMonitor: use new style options (pr#18209, Kefu Chai)
  • mon: osd/OSDMap.h: toss osd out if it has no more pending states (pr#19642, xie xingguo)
  • mon: paxos cleanup (pr#20078, huanwen ren)
  • mon/PGMap: let pg_string_state() return boost::optional<> (issue#21609, pr#18218, Kefu Chai)
  • mon/PGMap: use new-style options and cleanup (pr#18647, Kefu Chai)
  • mon: post-luminous cleanup (part 3 of ?) (pr#17607, Sage Weil)
  • mon: rate limit on health check update logging (issue#20888, pr#16942, John Spray)
  • mon: reenable timer to send digest when paxos is temporarily inactive (issue#22142, pr#19404, Jan Fajerski)
  • mon: remove health service (pr#20119, Chang Liu)
  • mon: remove_is_write_ready() (pr#19191, Kefu Chai)
  • mon: remove pre-luminous compat cruft (2 of many) (pr#17322, Sage Weil)
  • mon: remove unused waiting_for_commit (pr#18617, Kefu Chai)
  • mon: return directly after health_events_cleanup (pr#16964, wang yang)
  • mon: revert mds metadata argument name change (issue#22527, pr#19926, Patrick Donnelly)
  • mon: show feature flags when printing MonSession (pr#17535, Paul Emmerich)
  • mon: some cleanup (pr#17067, huanwen ren)
  • mon,tests: vstart: set osd_pool_default_erasure_code_profile in initial ceph.conf (pr#21008, Mykola Golub)
  • mon: update get_store_prefixes implementations (issue#21534, pr#17940, John Spray, huanwen ren)
  • mon: update PaxosService::cached_first_committed in PaxosService::maybe_trim() (issue#11332, pr#19397, Xuehan Xu, yupeng chen)
  • mon: use ceph_clock_now if message is self-generated (pr#17311, huangjun)
  • mon: warn about using osd new instead of osd create (issue#21023, pr#17242, Neha Ojha)
  • msg/async/AsyncConnection: remove legacy feature case handle (pr#18469, Haomai Wang)
  • msg/async: avoid referencing the temporary string (pr#20640, Kefu Chai)
  • msg/async: batch handle msg_iovlen (pr#18415, Jianpeng Ma)
  • msg/async/dpdk: remove xsky copyright and LGPL copying (pr#21121, Kefu Chai)
  • msg/async/EventKqueue: assert on OOM (pr#21488, Kefu Chai)
  • msg/async: fix ms_dpdk_coremask and ms_dpdk_coremask conflict (pr#18678, chunmei)
  • msg/async:fix the incoming parameter type of EventCenter::process_events() (pr#20607, shangfufei)
  • msg/async misc cleanup (pr#18531, Jianpeng Ma)
  • msg/async: misc cleanup (pr#18575, Jianpeng Ma)
  • msg/async/rdma: a tiny typo fix (pr#18660, Yan Lei)
  • msg/async/rdma: fix a coredump introduced by PR #18053 (pr#18204, Yan Lei)
  • msg/async/rdma: fix a potential coredump when handling tx_buffers under heavy RDMA (pr#18036, Yan Lei)
  • msg/async/rdma: fixes crash for multi rados client within one process (pr#16981, Alex Mikheev, Haomai Wang, Adir Lev)
  • msg/async/rdma: fix Tx buffer leakage that can introduce “heartbeat no reply” (pr#18053, Yan Lei)
  • msg/async/rdma: refactor rx buffer pool allocator (pr#17018, Alex Mikheev)
  • msg/async/rdma: unnecessary reinitiliazation of an iterator (pr#18190, JustL)
  • msg/async: size of EventCenter::file_events should be greater than fd (issue#23253, pr#20764, Yupeng Chen)
  • msg/async: use bitset<> to do the popcnt (pr#18681, Kefu Chai)
  • msg/async: use device before checking (pr#19738, Xiaoyan Li)
  • msg: drop duplicate include (pr#19623, /bin/bash)
  • msg: drop the unnecessary polling_stop() (pr#17079, Jos Collin)
  • msg: Initialize lkey,bound,port_cnt,num_chunk,gid_idx (pr#17797, Amit Kumar)
  • msg: Initializing class members in module msg (pr#17568, Amit Kumar)
  • msg: reimplement sigpipe blocking (pr#18105, Greg Farnum)
  • msg: remove the ),it’s redundant (pr#17544, linxuhua)
  • msg: resurrect support for !CEPH_FEATURE_MSG_AUTH (pr#19044, Ilya Dryomov)
  • msgr: Optimization for connection establishment (pr#16006, shangfufei)
  • msg/simple: pass a char for reading from shutdown_rd_fd (pr#19094, Kefu Chai)
  • NVMDevice: fix issued caused by #17002 (pr#17112, Ziye Yang)
  • objclass-sdk: expose __cls_init() to the world (pr#21581, Kefu Chai)
  • objecter: minor cleanups (pr#19994, runsisi)
  • os/bluestore/bluestore_tool: Move redundant code into one method (pr#19160, Shinobu Kinjo)
  • os/bluestore: implement BlueRocksEnv::AreFilesSame() (issue#21842, pr#18392, Kefu Chai)
  • os/bluestore: simplify and fix SharedBlob::put() (issue#24211, pr#22170, Sage Weil)
  • osd: additional protection for out-of-bounds EC reads (issue#21629, pr#18088, Jason Dillaman)
  • osd: add multiple objecter finishers (pr#16521, Myoungwon Oh)
  • osd: add num_object_manifest (pr#20690, Myoungwon Oh)
  • osd: add numpg_removing metric (pr#18450, Sage Weil)
  • osd: add processed_subop_count for cls_cxx_subop_version() (issue#21964, pr#18610, Casey Bodley)
  • osd: add scrub week day constraint (pr#18368, kungf)
  • osd: adjust osd_min_pg_log_entries (issue#21026, pr#17075, J. Eric Ivancich)
  • osd: allow FULL_TRY after failsafe (pr#17177, Pan Liu)
  • osd: allow PG recovery scheduling preemption (pr#17839, Sage Weil)
  • osd: async recovery (pr#19811, Neha Ojha)
  • osd: avoid encoding the same log entries repeatedly for different peers (pr#20201, Jianpeng Ma)
  • osd: avoid the config’s get_val() overhead on the read path (pr#20217, Radoslaw Zarzynski)
  • osd: avoid unnecessary ref-counting across PrimaryLogPG::get_rw_locks (pr#21028, Radoslaw Zarzynski)
  • osd: be more precise about our asserts and cases when rebuilding missing sets (issue#20985, pr#17000, Greg Farnum)
  • osd: bring in dmclock library changes (pr#16755, J. Eric Ivancich)
  • osd: bring in latest dmclock library updates (pr#17997, J. Eric Ivancich)
  • osd: cap snaptrimq_len at 2^32 (pr#21107, Kefu Chai)
  • osd: change log level when withholding pg creation (issue#22440, pr#20167, Dan van der Ster)
  • osd: change op delayed state to ‘waiting for scrub’ (pr#19295, kungf)
  • osd: Change shard digests to hex like object info digests (pr#21362, David Zafman)
  • osd: change the conditional in _update_calc_stats (pr#13383, Zhiqiang Wang)
  • osd: check feature bits when encoding objectstore_perf_stat_t (pr#20378, Kefu Chai)
  • osd: clean up dup index logic; maintain index flag logic in fewer places (pr#16829, J. Eric Ivancich)
  • osd: clean up pre-luminous compat cruft (part 1 of many) (pr#17247, Sage Weil)
  • osd: cleanups (pr#17753, Kefu Chai)
  • osdc/Objecter: using coarse_mono instead (pr#18473, Haomai Wang)
  • osdc/Objector: use std::shared_mutex instead of boost::shared_mutex (issue#23910, pr#21702, Abhishek Lekshmanan)
  • osd: correct several spell errors in comments (pr#21064, songweibin)
  • osdc: Remove a bit too redundant public label (pr#19466, Shinobu Kinjo)
  • osdc: self-managed snapshot helper should catch decode exception (issue#24103, issue#24000, pr#21958, Jason Dillaman)
  • osd: debug dispatch_context cases where queries not sent (pr#20917, Sage Weil)
  • osd: Deleting dead code PrimaryLogPG.cc (pr#17339, Amit Kumar)
  • osd: don’t crash on empty snapset (issue#23851, pr#21058, Mykola Golub, Igor Fedotov)
  • osd: Don’t include same header twice (pr#18319, Shinobu Kinjo)
  • osd: Don’t initialze pointers by NULL or 0 (pr#18311, Shinobu Kinjo)
  • osd: don’t memcpy hobject_t in PrimaryLogPG::close_op_ctx() (pr#21029, Radoslaw Zarzynski)
  • osd: don’t process ostream strings when not debugging (pr#20298, Mark Nelson)
  • osd: drop redundant comment (pr#20347, songweibin)
  • osd: Drop the unused code in OSD::_collect_metadata (pr#17131, Luo Kexue)
  • osd: drop unused osd_disk_tp related options (pr#21339, Gu Zhongyan)
  • osd: eliminate ineffective container operations (pr#19099, Igor Fedotov)
  • osd: enumerate device names in a simple way (pr#18453, Sage Weil)
  • osd: exit(1) directly without lock if init fails (pr#16647, Kefu Chai)
  • osd: fast dispatch of peering events and pg_map + osd sharded wq refactor (pr#19973, Sage Weil)
  • osd: fine-grained statistics of logical object space usage (pr#15199, xie xingguo)
  • osd: Fix assert when checking missing version (issue#21218, pr#20410, David Zafman)
  • osd: fix a valgrind issue (conditional jump depends on uninitialized value) (issue#22641, pr#19874, Myoungwon Oh)
  • osd: fix bug which cause can’t erase OSDShardPGSlot (pr#21771, Jianpeng Ma)
  • osd: fix build_initial_pg_history (issue#21203, pr#17423, w11979, Sage Weil)
  • osd: fix crash caused by divide by zero in heartbeat code (pr#21373, Piotr Dałek)
  • osd: fix dpdk memzon mz_name setting issue (pr#19809, chunmei Liu)
  • osd: fix dpdk runtime issue based on spdk/dpdk libarary (pr#19559, chunmei Liu)
  • osd: fix dpdk worker references issue (pr#19886, chunmei Liu)
  • osd: Fixes for osd_scrub_during_recovery handling (issue#18206, pr#17039, David Zafman)
  • osd: fix out of order caused by letting old msg from down osd be processed (issue#22570, pr#19796, Mingxin Liu)
  • osd: fix _process handling for pg vs slot race (pr#21745, Sage Weil)
  • osd: fix recovery reservation bugs, and implement remote reservation preemption (pr#18485, Sage Weil)
  • osd: fix replica/backfill target handling of REJECT (issue#21613, pr#18070, Sage Weil)
  • osd: fix reqid assignment for reply messages in OpRequest (pr#17060, Yingxin Cheng)
  • osd: fix s390x build failure (issue#23238, pr#20969, Nathan Cutler)
  • osd: fix typos and some cleanups (pr#19211, Enming Zhang)
  • osd: fix unordered read bug (for chunked object) (issue#22369, pr#19464, Myoungwon Oh)
  • osd: fix waiting_for_peered vs flushing (issue#21407, pr#17759, Sage Weil)
  • osd: flush operations for chunked objects (pr#19294, Myoungwon Oh)
  • osd: generalize queueing and lock interface for OpWq (pr#16030, Myoungwon Oh, Kefu Chai, Samuel Just)
  • osd: get loadavg per cpu for scrub load threshold check (pr#17718, kungf)
  • osd: get rid off extent map in object_info (pr#19616, Igor Fedotov)
  • osd: hold lock while accessing recovery_needs_sleep (issue#21566, pr#18022, Neha Ojha)
  • osd: Improve recovery stat handling by using peer_missing and missing_loc info (issue#22837, pr#20220, Sage Weil, David Zafman)
  • osd: Improve size scrub error handling and ignore system attrs in xattr checking (issue#20243, issue#18836, pr#16407, David Zafman)
  • osd: include front_iface+back_iface in metadata (issue#20956, pr#16941, John Spray)
  • osd: Initialization of data members (pr#17691, Amit Kumar)
  • osd: Initialization of pointer cls (pr#17115, amitkuma)
  • osd: Initializing start_offset,last_offset,offset (pr#19333, Amit Kumar)
  • osd: initial minimal efforts to clean up PG interface (pr#17708, Sage Weil)
  • osd: introduce sub-chunks to erasure code plugin interface (issue#19278, pr#15193, Myna Vajha)
  • osd: kill snapdirs (pr#17579, Sage Weil)
  • osd: Make dmclock’s anticipation timeout be configurable (pr#18827, Taewoong Kim)
  • osd: make operations on ReplicatedBackend::in_progress_ops more effective (pr#19035, Igor Fedotov)
  • osd: make PG::*Force* event structs public (pr#21312, Willem Jan Withagen)
  • osd: make scrub no deadline when max interval is zero (pr#18354, kungf)
  • osd: make scrub right now when pg stats_invalid is true (pr#17884, kungf)
  • osd: make scrub wait for ec read/modify/writes to apply (issue#23339, pr#20944, Sage Weil)
  • osd: make snapmapper warn+clean up instead of assert (issue#22752, pr#20040, Sage Weil)
  • osd: make stat_bytes and stat_bytes_used counters PRIO_USEFUL (issue#21981, pr#18637, Yao Zongyou)
  • osd: make the PG’s SORTBITWISE assert a more generous shutdown (issue#20416, pr#18047, Greg Farnum)
  • osd: Making use of find to reduce computational complexity (pr#19732, Shinobu Kinjo)
  • osd: migrate PGLOG_* macros to constexpr (issue#20811, pr#19352, Jesse Williamson)
  • osd: minor optimizations for op wq (pr#17704, Sage Weil, J. Eric Ivancich)
  • osd: min_pg_log_entries == max == pg_log_dups_tracked (pr#20394, Sage Weil)
  • osd: misc cleanups (pr#17430, songweibin)
  • osd: miscellaneous cleanups (pr#21431, songweibin)
  • osd: more debugging for snapmapper bug (issue#21557, pr#19366, Sage Weil)
  • osd: object added to missing set for backfill, but is not in recovering, error! (issue#18162, pr#18145, David Zafman)
  • osd: only exit if *latest* map(s) say we are destroyed (issue#22673, pr#19988, Sage Weil)
  • osd: Only scan for omap corruption once (issue#21328, pr#17705, David Zafman)
  • osd,os,io: Initializing C_ProxyChunkRead members,queue,request (pr#19336, amitkuma)
  • osd: pass ops_blocked_by_scrub() to requeue_scrub() (pr#20319, Kefu Chai)
  • osd: pass pool options to ObjectStore on pg create (issue#22419, pr#19486, Sage Weil)
  • osd/PG: fix clang build vs private state events (pr#18217, Sage Weil)
  • osd/PG: handle flushed event directly (pr#19441, wumingqiao)
  • osd/PrimaryLogPG: derr when object size becomes over osd_max_object_size (pr#19049, Shinobu Kinjo)
  • osd: process _scan_snaps() with all snapshots with head (issue#22881, issue#23909, pr#21546, David Zafman)
  • osd: publish osdmap to OSDService before starting wq threads (issue#21977, pr#21623, Sage Weil)
  • osd: pull latest dmclock subtree (pr#19345, J. Eric Ivancich)
  • osd: put peering events in main sharded wq (pr#18752, Sage Weil)
  • osd: put pg removal in op_wq (pr#19433, Sage Weil)
  • osd: reduce all_info map find to get primary (pr#19425, kungf)
  • osd: refcount for manifest object (redirect, chunked) (pr#19935, Myoungwon Oh)
  • osd: remove cost from mclock op queues; cost not handled well in dmclock (pr#21428, J. Eric Ivancich)
  • osd: Remove double space (pr#19296, Shinobu Kinjo)
  • osd: remove duplicated “commit_queued_for_journal_write” in OpTracker (issue#23440, pr#21018, ashitakasam)
  • osd: remove duplicated function ec_pool in pg_pool_t (pr#18059, Chang Liu)
  • osd: Remove redundant local variable declaration (pr#19812, Shinobu Kinjo)
  • osd: Remove unnecessary headers (pr#19735, Shinobu Kinjo)
  • osd: remove unused ReplicatedBackend::objects_read_async() (pr#18779, Kefu Chai)
  • osd: remove unused variable in do_proxy_write (pr#17391, Myoungwon Oh)
  • osd: replace mclock subop opclass w/ rep_op opclass; combine duplicated code (pr#18194, J. Eric Ivancich)
  • osd: replace vectors_equal() with operator==(vector<>, vector<>) (pr#18064, Kefu Chai)
  • osd: request new map from PG when needed (issue#21428, pr#17795, Josh Durgin)
  • osd: resend osd_pgtemp if it’s not acked (issue#23610, pr#21310, Kefu Chai)
  • osd: Revert use of dmclock message feature bit since not yet finalized (pr#21398, J. Eric Ivancich)
  • osd,rgw,librbd: SCA fixes (pr#18495, Danny Al-Gaaf)
  • osd: set min_version to newest version in maybe_force_recovery (pr#17752, Xinze Chi)
  • osd: Sign in early SIGHUP signal (issue#22746, pr#19958, huanwen ren)
  • osd: silence maybe-uninitialized false positives (pr#19820, Yao Zongyou)
  • osd: silence warnings from -Wsign-compare (pr#17872, Jos Collin)
  • osd: skip dumping logical devices (pr#20740, songweibin)
  • osd: speed up get_key_name (issue#21026, pr#17071, J. Eric Ivancich)
  • osd: s/random_shuffle()/shuffle()/ (pr#19872, Willem Jan Withagen, Kefu Chai, Greg Farnum)
  • osd: subscribe osdmaps if any pending pgs (issue#22113, pr#18916, Kefu Chai)
  • osd: subscribe to new osdmap while waiting_for_healthy (issue#21121, pr#17244, Sage Weil)
  • osd: support class method whitelisting within caps (pr#19786, Jason Dillaman)
  • osd: treat successful and erroroneous writes the same for log trimming (issue#22050, pr#20827, Josh Durgin)
  • osd: two cleanups (pr#20830, songweibin)
  • osd: update dmclock library w git subtree pull (pr#17737, J. Eric Ivancich)
  • osd: update info only if new_interval (pr#17437, Kefu Chai)
  • osd: update store with options after pg is created (issue#22419, pr#20044, Kefu Chai)
  • osd: use dmclock library client_info_f function dynamically (pr#17063, bspark)
  • osd: use existing osd_required variable for messenger policy (pr#20223, Yan Jun)
  • osd: use prefix increment for non trivial iterator (pr#19097, Kefu Chai)
  • osd: Use specializations, typedefs instead (pr#19354, Shinobu Kinjo)
  • osd: Warn about objects with too many omap entries (pr#16332, Brad Hubbard)
  • os/filestore/HashIndex.h: fixed a typo in comment (pr#17685, yaoguotao)
  • os: Initializing uninitialized members aio_info (pr#17066, amitkuma)
  • os: Removing dead code from LFNIndex.cc (pr#17297, Amit Kumar)
  • prometheus: Handle the TIME perf counter type metrics (pr#21749, Boris Ranto)
  • pybind: add return note in rbd.pyx (pr#21768, Zheng Yin)
  • pybind/ceph_daemon: expand the order of magnitude of (issue#23962, pr#21836, Guan yunfei)
  • pybind: fix chart size become bigger when refresh (issue#20746, pr#16857, Yixing Yan)
  • pybind: mgr/dashboard: fix rbd’s pool sub menu (pr#16774, yanyx)
  • pybind,rbd: pybind/rbd: support open the image by image_id (pr#19361, songweibin)
  • pybind: remove unused get_ceph_version() (pr#17727, Kefu Chai)
  • qa: add cbt repo parameter (pr#18543, Neha Ojha)
  • qa: Add cephmetrics suite (pr#18451, Zack Cerza)
  • qa: add upgrade/luminous-x suite (pr#17160, Yuri Weinstein)
  • qa/crontab: run the perf-basic suite every day (pr#21252, Neha Ojha)
  • qa: Decreased amount of jobs on master, kraken, luminous runs (pr#17069, Yuri Weinstein)
  • qa: install collectl with cbt task (pr#19324, Neha Ojha)
  • qa: mimic-dev1 backports to avoid trusty nodes (pr#19600, Kefu Chai)
  • qa: preserve cbt task results (pr#19364, Neha Ojha)
  • qa: qa/ceph-disk: enlarge the simulated SCSI disk (issue#22136, pr#19199, Kefu Chai)
  • qa/suites/perf-basic: add desc regarding test machines (pr#21183, Neha Ojha)
  • qa/suites/rados/multimon/tasks/mon_lock_with_skew: whitelist PG (pr#17004, Sage Weil)
  • qa/suites/rados/perf: add optimized settings (pr#17786, Neha Ojha)
  • qa/suites/rados/perf: add workloads (pr#18573, Neha Ojha)
  • qa/suites/rados/verify/validater/valgrind: whitelist PG (pr#17005, Sage Weil)
  • qa/suites/upgrade/jewel-x/parallel: tolerate laggy mgr (pr#17227, Sage Weil)
  • qa/suites/upgrade/kraken-x: fixes (pr#16881, Sage Weil)
  • qa/suites/upgrade/luminous-x fixes (pr#22101, Sage Weil)
  • qa/tests - Added options to use both cases: mon.a and installer.0 (pr#19745, Yuri Weinstein)
  • qa/tests - Fixed typo in crontab entry (pr#21482, Yuri Weinstein)
  • qa/tests: fixed typo (pr#21728, Yuri Weinstein)
  • qa/tests - minor clean ups and made perf-suite run 3 times, so we can… (pr#21309, Yuri Weinstein)
  • qa/tests - one more typo fixed :( (pr#21483, Yuri Weinstein)
  • qa/tests: removed rest suite from the mix (pr#21743, Yuri Weinstein)
  • qa: wait longer for osd to flush pg stats (issue#24321, pr#22288, Kefu Chai)
  • qa/workunits/ceph-disk: –no-mon-config (pr#21956, Kefu Chai)
  • rados: make ceph_perf_msgr_client work for multiple jobs (issue#22103, pr#18877, Jeegn Chen)
  • rbd: add deep cp CLI method (pr#19996, songweibin)
  • rbd: add group rename methods (issue#22981, pr#20577, songweibin)
  • rbd: add notrim option to rbd map (pr#21056, Hitoshi Kamei)
  • rbd: add parent info when moving child into trash bin (pr#19280, songweibin)
  • rbd: adjusted “lock list” JSON and XML formatted output (pr#19900, Jason Dillaman)
  • rbd: adjusted “showmapped” JSON and XML formatted output (pr#19937, Mykola Golub)
  • rbd: allow remove all unprotected snapshots (issue#23126, pr#20608, songweibin)
  • rbd: allow trash rm/purge when pool quota is full used (pr#20697, songweibin)
  • rbd: backport of mimic bug fixes (issue#24009, issue#24008, pr#21930, Jason Dillaman)
  • rbd: check if an image is already mapped before rbd map (issue#20580, pr#16517, Jing Li)
  • rbd: children list should support snapshot id optional (issue#23399, pr#20966, Jason Dillaman)
  • rbd: cleanup handling of IEC byte units (pr#21564, Jason Dillaman)
  • rbd: clean up warnings when mirror commands used on non-setup pool (issue#21319, pr#17636, Jason Dillaman)
  • rbd: cls/journal: ensure tags are properly expired (issue#21960, pr#18604, Jason Dillaman)
  • rbd: cls/journal: fixed possible infinite loop in expire_tags (issue#21956, pr#18592, Jason Dillaman)
  • rbd: cls/journal: possible infinite loop within tag_list class method (issue#21771, pr#18270, Jason Dillaman)
  • rbd: cls/rbd: group_image_list incorrectly flagged as RW (issue#23388, pr#20939, Jason Dillaman)
  • rbd: cls/rbd: metadata_list not honoring max_return parameter (issue#21247, pr#17499, Jason Dillaman)
  • rbd: cls/rbd: Silence gcc7 maybe-uninitialized warning (pr#18504, Brad Hubbard)
  • rbd: common/options,librbd/Utils: refactor RBD feature validation (pr#20014, Sage Weil)
  • rbd: disk usage on empty pool no longer returns an error message (issue#22200, pr#19045, Jason Dillaman)
  • rbd: do not show title if there is no group snapshot (pr#20311, songweibin)
  • rbd: don’t overwrite the error code from the remove action (pr#20481, Jason Dillaman)
  • rbd: drop unnecessary using declaration, etc (pr#19005, Shinobu Kinjo)
  • rbd: eager-thick provisioning support (pr#18317, Hitoshi Kamei)
  • rbd: export/import image-meta when we export/import an image (pr#17134, PCzhangPC)
  • rbd: filter out UserSnapshotNamespace in do_disk_usage (pr#20532, songweibin)
  • rbd: fix crash during map when “rw” option is specified (issue#21808, pr#18313, Peter Keresztes Schmidt)
  • rbd: fix logically dead code in function list_process_image (pr#16971, Luo Kexue)
  • rbd: fix rbd children listing when child is in trash (issue#21893, pr#18483, songweibin)
  • rbd: fix thread_offsets calculation of rbd bench (pr#20590, Hitoshi Kamei)
  • rbd: group misc cleanup and update rbd man page (pr#20199, songweibin)
  • rbd: group snapshot rename (pr#12431, Victor Denisov)
  • rbd: implement image qos in tokenbucket algorithm (pr#17032, Dongsheng Yang)
  • rbd: import with option –export-format 2 fails to protect snapshot (issue#23038, pr#20613, songweibin)
  • rbd: improve ‘import-diff’ corrupt input error messages (issue#18844, pr#21249, Jason Dillaman)
  • rbd: Initializing m_finalize_ctx (pr#17563, Amit Kumar)
  • rbd: introduce commands of “image-meta ls/rm” (pr#16591, PCzhangPC)
  • rbd: journal: limit number of appends sent in one librados op (issue#23526, pr#21157, Mykola Golub)
  • rbd: journal: trivial cleanup (pr#19317, Shinobu Kinjo)
  • rbd: krbd: include sys/sysmacros.h for major, minor and makedev (pr#20773, Ilya Dryomov)
  • rbd: krbd: rewrite “already mapped” code (pr#17638, Ilya Dryomov)
  • rbd: librados/snap_set_diff: don’t assert on empty snapset (pr#20648, Mykola Golub)
  • rbd: librbd: create+truncate for whole-object layered discards (issue#23285, pr#20809, Ilya Dryomov)
  • rbd: librbd: make rename request complete with filtered code (issue#23068, pr#20507, Mykola Golub)
  • rbd: librbd misc cleanup (pr#18419, Jianpeng Ma)
  • rbd: librbd: skip head object map update when deep copying object beyond image size (pr#21586, Mykola Golub)
  • rbd: librbd: sync flush should re-use existing async flush logic (pr#18403, Jason Dillaman)
  • rbd: librbd,test: address coverity false positives (pr#17825, Amit Kumar)
  • rbd: mimic: librbd: deep copy optionally support flattening cloned image (issue#22787, pr#22038, Mykola Golub)
  • rbd: mimic: rbd-mirror: optionally support active/active replication (pr#22105, Jason Dillaman)
  • rbd: mimic: rbd-mirror: potential deadlock when running asok ‘flush’ command (issue#24141, pr#22039, Mykola Golub)
  • rbd-mirror: additional thrasher testing (pr#21697, Jason Dillaman)
  • rbd-mirror: clean up spurious error log messages (issue#21961, pr#18601, Jason Dillaman)
  • rbd-mirror: cluster watcher should ensure it has latest OSD map (issue#22461, pr#19550, Jason Dillaman)
  • rbd-mirror: ensure unique service daemon name is utilized (pr#19492, Jason Dillaman)
  • rbd-mirror: fix potential infinite loop when formatting status message (issue#22932, pr#20349, Mykola Golub)
  • rbd-mirror: forced promotion can result in incorrect status (issue#21559, pr#17979, Jason Dillaman)
  • rbd-mirror: ImageMap memory leak fixes (pr#19163, Venky Shankar)
  • rbd-mirror: Improve data pool selection when creating images (pr#18006, Adam Wolfe Gordon)
  • rbd-mirror: integrate image map policy as incremental step to active-active (pr#21300, Jason Dillaman)
  • rbd-mirror: introduce basic image mapping policy (issue#18786, pr#15691, Venky Shankar)
  • rbd-mirror: missing lock when re-sending update_sync_point (pr#19011, Mykola Golub)
  • rbd-mirror: persist image map timestamp (pr#19338, Venky Shankar)
  • rbd-mirror: primary image should register in remote, non-primary image’s journal (issue#21561, pr#18136, Jason Dillaman)
  • rbd-mirror: properly translate remote tag mirror uuid for local mirror (issue#23876, pr#21657, Jason Dillaman)
  • rbd-mirror: removed dedicated thread from image deleter (issue#15322, pr#19000, Jason Dillaman)
  • rbd-mirror: rename asok hook to match image name when not replaying (issue#23888, pr#21682, Jason Dillaman)
  • rbd-mirror: rollback state transitions in image policy (pr#19577, Venky Shankar)
  • rbd-mirror: Set the data pool correctly when creating images (issue#20567, pr#17073, Adam Wolfe Gordon)
  • rbd-mirror: simplify notifications for image assignment (issue#15764, pr#16642, Jason Dillaman)
  • rbd-mirror: strip environment/CLI overrides for remote cluster (issue#21894, pr#18490, Jason Dillaman)
  • rbd-mirror: support deferred deletions of mirrored images (pr#19536, Jason Dillaman)
  • rbd-mirror: sync image metadata when transfering remote image (issue#21535, pr#18026, Jason Dillaman)
  • rbd-mirror: track images in policy map in support of A/A (issue#18786, pr#15788, Venky Shankar)
  • rbd-mirror: update asok hook name on image rename (issue#20860, pr#16998, Mykola Golub)
  • rbd-mirror: use next transition state to check transition completeness (pr#18969, Venky Shankar)
  • rbd-nbd: allow to unmap by image or snap spec (pr#19666, Mykola Golub)
  • rbd-nbd: bug fix when running in container (issue#22012, issue#22011, pr#18663, Li Wang)
  • rbd-nbd: certain kernels may not discover resized block devices (issue#22131, pr#18947, Jason Dillaman)
  • rbd-nbd: cleanup for NBDServer shut down (pr#17283, Pan Liu)
  • rbd-nbd: fix ebusy when do map (issue#23528, pr#21142, Li Wang)
  • rbd-nbd: fix generic option issue (issue#20426, pr#17375, Pan Liu)
  • rbd-nbd: output format support for list-mapped command (pr#19704, Mykola Golub)
  • rbd-nbd: support optionally setting device timeout (issue#22333, pr#19436, Mykola Golub)
  • rbd: null check before pool_name use (pr#18790, Amit Kumar)
  • rbd: output notifyOp request name when watching (pr#20551, shun-s)
  • rbd: parallelize “rbd ls -l” (pr#15579, Piotr Dałek)
  • rbd: pool_percent_used should not divided by 100 (pr#20795, songweibin)
  • rbd: properly pass ceph global command line args to subprocess (pr#19821, Mykola Golub)
  • rbd: pybind/rbd: add deep_copy method (pr#19406, Mykola Golub)
  • rbd: pybind/rbd: fix metadata functions error handling (issue#22306, pr#19337, Mykola Golub)
  • rbd: python bindings fixes and improvements (issue#23609, pr#21304, Ricardo Dias)
  • rbd: rbd-ggate: fix parsing ceph global options (pr#19822, Mykola Golub)
  • rbd: rbd-ggate: fix syntax error (pr#19919, Willem Jan Withagen)
  • rbd: rbd-ggate: make list command produce valid xml format output (pr#19823, Mykola Golub)
  • rbd: rbd-ggate: small fixes and improvements (pr#19679, Mykola Golub)
  • rbd: rbd-ggate: tool to map images on FreeBSD via GEOM Gate (pr#15339, Mykola Golub)
  • rbd: rbd:introduce rbd bench rw(for read and write mix) test (pr#17461, PCzhangPC)
  • rbd: rbd: set a default value for options in nbd map (pr#20529, songweibin)
  • rbd: replace positional_path parameter with arg_index in get_path() (pr#19722, songweibin)
  • rbd: replace trash delay option, add rbd trash purge command (pr#18323, Theofilos Mouratidis)
  • rbd: resource agent needs to be executable (issue#22980, issue#22362, pr#20397, Tim Bishop)
  • rbd:rm unnecessary conversion from string to char* in image-meta function (pr#17184, PCzhangPC)
  • rbd: show read:write proportion in the infomation of readwrite bench test (pr#18249, PCzhangPC)
  • rbd: snap limit should’t be set smaller than the number of existing snaps (pr#16597, PCzhangPC)
  • rbd: support cloning an image from a non-primary snapshot (issue#18480, pr#19724, Jason Dillaman)
  • rbd: support iterating over metadata items when listing (issue#21179, pr#17532, Jason Dillaman)
  • rbd: support lock_timeout in rbd mapping (pr#21344, Dongsheng Yang)
  • rbd: support osd_request_timeout in rbd map command (issue#23073, pr#20792, Dongsheng Yang)
  • rbd: switched from legacy to new-style configuration options (issue#20737, pr#16737, Jason Dillaman)
  • rbd,tests: qa: additional krbd discard test cases (pr#20499, Ilya Dryomov)
  • rbd,tests: qa: fix POOL_APP_NOT_ENABLED warning in krbd:unmap suite (pr#16966, Ilya Dryomov)
  • rbd,tests: qa: introduce rbd-mirror thrasher to existing tests (issue#18753, pr#21541, Jason Dillaman)
  • rbd,tests: qa: krbd_exclusive_option.sh: add lock_timeout test case (pr#21522, Ilya Dryomov)
  • rbd,tests: qa: krbd_fallocate.sh: add notrim test case (pr#21513, Ilya Dryomov)
  • rbd,tests: qa: krbd huge-image test (pr#20692, Ilya Dryomov)
  • rbd,tests: qa: krbd latest-osdmap-on-map test (pr#20591, Ilya Dryomov)
  • rbd,tests: qa: krbd msgr-segments test (pr#20714, Ilya Dryomov)
  • rbd,tests: qa: krbd parent-overlap test (pr#20721, Ilya Dryomov)
  • rbd,tests: qa: krbd whole-object-discard test (pr#20750, Ilya Dryomov)
  • rbd,tests: qa/suites/krbd: add krbd BLKROSET test (pr#18652, Ilya Dryomov)
  • rbd,tests: qa/suites/krbd: enable generic/050 and generic/448 (pr#18795, Ilya Dryomov)
  • rbd,tests: qa/suites/krbd: enable xfstests blockdev tests (pr#17621, Ilya Dryomov)
  • rbd,tests: qa/suites/krbd: exclude shared/298 (pr#17971, Ilya Dryomov)
  • rbd,tests: qa/suites/krbd: rbd_xfstests job overhaul (pr#17346, Ilya Dryomov)
  • rbd,tests: qa/suites/rbd: fewer socket failures (pr#19617, Sage Weil)
  • rbd,tests: qa/suites/rbd: miscellaneous test fixes (issue#21251, pr#17504, Jason Dillaman)
  • rbd,tests: qa/suites/rbd: segregated v1 image format tests (issue#22738, pr#20729, Jason Dillaman)
  • rbd,tests: qa/suites/rbd: set qemu task time_wait param (pr#21131, Mykola Golub)
  • rbd,tests: qa/tasks/cram: include /usr/sbin in the PATH for all commands (pr#18793, Ilya Dryomov)
  • rbd,tests: qa/tasks/rbd: run all xfstests runs to completion (pr#18583, Ilya Dryomov)
  • rbd,tests: qa/workunits/rbd: fix cli_generic test_purge for rbd default format 1 (pr#20389, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: fixed variable name for resync image id (issue#21663, pr#18097, Jason Dillaman)
  • rbd,tests: qa/workunits/rbd: fix issues within permissions test (issue#23043, pr#20491, Jason Dillaman)
  • rbd,tests: qa/workunits/rbd: pool create may fail for small cluster (pr#18067, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: potential race in mirror disconnect test (issue#23938, pr#21733, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: relax greps to support upgrade formatting change (issue#21181, pr#17559, Jason Dillaman)
  • rbd,tests: qa/workunits/rbd: remove sanity check in journal.sh test (pr#20490, Jason Dillaman)
  • rbd,tests: qa/workunits/rbd: remove sanity check in test_admin_socket.sh (pr#21116, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: remove “trash purge –threshold” test (issue#22803, pr#20170, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: simplify split-brain test to avoid potential race (issue#22485, pr#19604, Jason Dillaman)
  • rbd,tests: qa/workunits/rbd: switch devstack tempest to 17.2.0 tag (issue#22961, pr#20599, Jason Dillaman)
  • rbd,tests: qa/workunits/rbd: switch devstack to pike release (pr#20124, Jason Dillaman)
  • rbd,tests: qa/workunits/rbd: test data pool is mirrored correctly (pr#17062, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: unnecessary sleep after failed remove (pr#18619, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: use command line option to specify watcher asok (issue#20954, pr#16917, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: wait for demote status is propagated (pr#19073, Mykola Golub)
  • rbd,tests: qa/workunits/rbd: wait for status propagated only if daemon started (pr#19082, Mykola Golub)
  • rbd,tests: rbd/test: add snap protection test for ex/import (pr#20689, songweibin)
  • rbd,tests: stop.sh: use –no-mon-config when trying to unmap rbd devices (pr#21020, Mykola Golub)
  • rbd,tests: test: address coverity false positives (pr#17803, Amit Kumar)
  • rbd,tests: test/cls_rbd: mask newer feature bits to support upgrade tests (issue#21217, pr#17509, Jason Dillaman)
  • rbd,tests: test/librados_test_stub: always create copy of buffers passed to operation (pr#21074, Mykola Golub)
  • rbd,tests: test/librbd: added update_features RPC message to test_notify (issue#21936, pr#18561, Jason Dillaman)
  • rbd,tests: test/librbd: clean up for several mock function tests (pr#18952, Jason Dillaman)
  • rbd,tests: test/librbd: Do not instantiate TrimRequest template class (pr#19402, Boris Ranto)
  • rbd,tests: test/librbd: ensure OutOfOrder test has enough concurrent management ops (pr#21436, Mykola Golub)
  • rbd,tests: test/librbd: fix mock method macro of set_journal_policy (pr#17216, Yan Jun)
  • rbd,tests: test/librbd: fix race condition with OSD map refresh (issue#20918, pr#16877, Jason Dillaman)
  • rbd,tests: test/librbd: fix valgrind memory leak warning (pr#17187, Mykola Golub)
  • rbd,tests: test/librbd: initialize on_finish,locker,force,snap_id (pr#17800, Amit Kumar)
  • rbd,tests: test/librbd: make fsx build on non-linux platform (pr#16939, Mykola Golub)
  • rbd,tests: test/librbd: memory leak in recently added test (pr#18478, Mykola Golub)
  • rbd,tests: test/librbd: rbd-ggate mode for fsx (pr#19315, Mykola Golub)
  • rbd,tests: test/librbd: test metadata_set/remove is applied (pr#18288, Mykola Golub)
  • rbd,tests: test/librbd: TestMirroringWatcher unit tests should ignore duplicates (issue#21029, pr#17078, Jason Dillaman)
  • rbd,tests: test/librbd: utilize unique pool for cache tier testing (issue#11502, pr#20486, Jason Dillaman)
  • rbd,tests: test/librbd: valgrind warning in TestMockManagedLockBreakRequest.DeadLockOwner (pr#18940, Mykola Golub)
  • rbd,tests: test/pybind/rbd: skip test_deep_copy_clone if layering not enabled (pr#20295, Mykola Golub)
  • rbd,tests: test/rbd: cli_generic fails if v1 image format or deep-flatten disabled (issue#22950, pr#20364, songweibin)
  • rbd,tests: test/rbd_mirror: fix valgrind warnings in unittest (pr#19016, Mykola Golub)
  • rbd,tests: test/rbd-mirror: image map policy test (pr#19320, Venky Shankar)
  • rbd,tests: test/rbd-mirror: improve coverage for dead instance handling (pr#21403, Jason Dillaman)
  • rbd,tests: test/rbd_mirror: “use of uninitialised value” valgrind warning (pr#19437, Mykola Golub)
  • rbd,tools: rbd-fuse: make sure PATH_MAX is defined (pr#18615, Roberto Oliveira)
  • rbd,tools: rbd-replay: remove boost dependency (pr#21202, Kefu Chai)
  • rbd: tools/rbd: use steady clock in bencher (pr#20008, Mohamad Gebai)
  • rbd: ‘trash list –long’ will return a failure on non-cloned images (pr#19540, Jason Dillaman)
  • rbd: ‘trash ls -l’ will display column titles if existed non-USER trash image only (pr#21343, songweibin)
  • rbd: unified way to map images using different drivers (pr#19711, Mykola Golub)
  • rbd: use different logic to disturb thread’s offset in bench seq test (pr#17218, PCzhangPC)
  • Revert “ceph-fuse: Delete inode’s bufferhead was in Tx state would le… (pr#21976, “Yan, Zheng”)
  • Revert “msg/async/rdma: fix multi cephcontext confllicting” (pr#16980, Haomai Wang)
  • Revert “os/bluestore: compensate for bad freelistmanager size/blocks metadata” (pr#17275, Xie Xingguo)
  • rgw: ability to list bucket contents in unsorted order for efficiency (pr#21026, J. Eric Ivancich)
  • rgw: abort multipart if upload meta object doesn’t exist (pr#19918, fang yuxiang)
  • rgw: Access RGWConf through RGWEnv (pr#17432, Jos Collin)
  • rgw: add “Accept-Ranges” to response header of Swift API (issue#21554, pr#17967, Tone Zhang)
  • rgw: add a default redirect field for zones (pr#9571, Yehuda Sadeh)
  • rgw: add an option to clear all usage entries (pr#19322, Abhishek Lekshmanan)
  • rgw: add an option to recalculate user stats (issue#23335, pr#20853, Abhishek Lekshmanan)
  • rgw: add buffering filter to compression for fetch_remote_obj (issue#23547, pr#21479, Casey Bodley)
  • rgw: add cors header rule check in cors option request (issue#22002, pr#18556, yuliyang)
  • rgw: Add dynamic resharding documentation (issue#21553, pr#15941, Orit Wasserman)
  • rgw: add logs if get_data returns error in RGWPutObj::execute (pr#18642, Zhang Shaowen)
  • rgw: add metadata and data sync related cmd into radosgw-admin usage (pr#18921, lvshanchun)
  • rgw: add missing override in list_keys_init() (pr#17254, Jos Collin)
  • rgw: add radosgw-admin sync error trim to trim sync error log (pr#19854, fang yuxiang)
  • rgw: add reshard commands (issue#21617, pr#18180, Orit Wasserman)
  • rgw: address warnings due to incorrect format code (pr#18796, J. Eric Ivancich)
    rgw: Add retry_raced_bucket_write
  • rgw: add rewrite cmd and options into radosgw-admin usage and doc (pr#18918, Enming Zhang)
  • rgw: add ssl support to beast frontend (issue#22832, pr#20464, Casey Bodley)
  • rgw: add support for Swift’s per storage policy statistics (issue#17932, pr#12704, Radoslaw Zarzynski)
  • rgw: add support for Swift’s reversed account listings (issue#21148, pr#17320, Radoslaw Zarzynski)
  • rgw: add support for tagging and other conditionals in policy (pr#17094, Abhishek Lekshmanan)
  • rgw: add tail tag to track tail instance (issue#20234, pr#16145, Yehuda Sadeh)
  • rgw: add tenant to shard_id in RGWDeleteLC::execute() (pr#10460, Wei Qiaomiao)
  • rgw: add time skew check in function parse_v4_auth_header (issue#22418, pr#19476, Bingyin Zhang)
    rgw: Add try_refresh_bucket_info function
  • rgw: add xml output header in RGWCopyObj_ObjStore_S3 response msg (issue#22416, pr#19475, Enming Zhang)
  • rgw: adjust log format for lifecycle (pr#19576, Bingyin Zhang)
  • rgw: admin api - add ability to sync user stats from admin api (issue#21301, pr#17589, Nathan Johnson)
  • rgw: Admin API Support for bucket quota change (issue#21811, pr#18324, Jeegn Chen)
  • rgw: admin rest api shouldn’t return error when getting user’s stats if the user hasn’t create any bucket (pr#21551, Zhang Shaowen)
  • rgw: allow beast frontend to listen on specific IP address (issue#22778, pr#20000, Yuan Zhou)
  • rgw: Allow swift acls to be deleted (issue#22897, pr#20471, Marcus Watts)
  • rgw: avoid logging keystone revocation messages when not configured (issue#21400, pr#17775, Abhishek Lekshmanan)
  • rgw: aws4 auth supports PutBucketRequestPayment (issue#23803, pr#21569, Casey Bodley)
  • rgw: AWS v4 authorization work when INIT_MULTIPART is chunked (issue#22129, pr#18956, Jeegn Chen)
  • rgw: beast frontend can listen on multiple endpoints (issue#22779, pr#20188, Casey Bodley)
  • rgw: beast frontend no longer experimental (pr#21272, Casey Bodley)
  • rgw: Better ERANGE error message (issue#22351, pr#20023, Brad Hubbard)
  • rgw: break sending data-log list infinitely (issue#20951, pr#16926, fang.yuxiang)
  • rgw: bucket resharding should not update bucket ACL or user stats (issue#22742, issue#22124, pr#20038, Orit Wasserman)
  • rgw: Cache on the barrelhead (issue#22517, pr#19581, Adam C. Emerson)
  • rgw: Cache Register! (issue#22604, issue#22603, pr#20144, Adam C. Emerson)
  • rgw: can’t download object with range when compression enabled (issue#22852, pr#20226, fang yuxiang)
  • rgw: ceph-dencoder: add missing begin_iter & end_iter item for RGWObjManifest (pr#19509, wangsongbo)
  • rgw: ceph-dencoder: add support for cls_rgw_lc_obj_head (pr#18920, Yao Zongyou)
  • rgw: ceph-dencoder: add support for RGWLifecycleConfiguration (pr#18959, wangsongbo)
  • rgw: change ObjectCache::lru from deque back to list (issue#22560, pr#19768, Casey Bodley)
  • rgw: changes to support ragweed (pr#13644, Yehuda Sadeh)
  • rgw: Check bucket CORS operations in policy (issue#21578, pr#18000, Adam C. Emerson)
  • rgw: Check bucket GetBucketLocation in policy (issue#21582, pr#18002, Adam C. Emerson)
  • rgw: Check bucket Website operations in policy (issue#21597, pr#18024, Adam C. Emerson)
  • rgw: check going_down() when lifecycle processing (issue#22099, pr#18846, Yao Zongyou)
  • rgw: Check payment operations in policy (issue#21389, pr#17742, Adam C. Emerson)
  • rgw: check read_op.read return value in RGWRados::copy_obj_data (pr#18962, Enming Zhang)
  • rgw: civetweb fixes for v1.1 upgrade (pr#21123, Abhishek Lekshmanan)
  • rgw: clean code with helper function dump_header_if_nonempty (pr#18979, Xinying Song)
  • rgw: clean up and fix some bugs for encryption (issue#21581, pr#17882, Enming Zhang)
  • rgw: cleanup MIN macro with std::min (pr#17546, Jiaying Ren)
  • rgw: cleanup unused parameters in RGWRados::copy_obj_data (pr#18917, Enming Zhang)
  • rgw: cloud sync fixes (pr#21648, Yehuda Sadeh)
  • rgw: cls/log: cls_log_list always returns next marker (issue#20906, pr#17024, Casey Bodley)
  • rgw: cls/rgw: fix bi_log_iterate_entries return wrong truncated (issue#22737, pr#20021, Tianshan Qu)
  • rgw: cls/rgw: Initialization of uninitialized members (pr#16932, amitkuma)
  • rgw: cls/rgw: mtime in rgw_bucket_dir_entry_meta not really decoded (issue#22148, pr#18981, Yao Zongyou)
  • rgw: cls/rgw: remove unused variable bl (pr#19570, Yao Zongyou)
  • rgw: cls/rgw: trim all usage entries in cls_rgw (issue#22234, pr#19131, Abhishek Lekshmanan)
  • rgw: cls_rgw: use more effective container operations in get_obj_vals (pr#19272, Xinying Song)
  • rgw: comparison between signed and unsigned integer expressions (pr#21105, ashitakasam)
  • rgw: consolidate code that implements hashing algorithms (pr#18248, J. Eric Ivancich)
  • rgw: copy object add response error messages (pr#18291, Enming Zhang)
  • rgw: correct comment in function parse_credentials (pr#19275, Bingyin Zhang)
  • rgw: correct log output for metadata section name in RGWListBucketIndexesCR (pr#19508, Xinying Song)
  • rgw: Correct permission evaluation to allow only admin users to work with Roles (pr#20332, Pritha Srivastava)
  • rgw: correct typo refity to refit (pr#19064, Bingyin Zhang)
  • rgw: correct typo UNKOWN to UNKNOWN (pr#19273, Bingyin Zhang)
  • rgw: create sync-module instance when execute radosgw-admin data sync run (issue#22080, pr#18898, lvshanchun)
  • rgw: create sync-module instance when radosgw-admin sync run (pr#20611, lvshanchun)
  • rgw: curl* reuse and for debian, use openssl not gnutls (pr#20635, Marcus Watts)
  • rgw: Data encryption is not follow the AWS agreement (pr#15994, hechuang)
  • rgw: datalog list support –shard-id and –marker (pr#20649, Tianshan Qu)
  • rgw: data sync: set num_shards when building full maps (issue#22083, pr#18852, Abhishek Lekshmanan)
  • rgw: Delete to_string functions. stringify defined in include/stringify.h can provide the same feature (pr#18522, zhangwen)
  • rgw: disable dynamic resharding in multisite environment (issue#21725, pr#18184, Orit Wasserman)
  • rgw: do not reflect period if not current (issue#22844, pr#20212, Tianshan Qu)
  • rgw: do not update all gateway caches upon creation of system obj w/ exclusive flag (pr#19384, J. Eric Ivancich)
  • rgw: don’t change rados object’s mtime when update olh (issue#21743, pr#18214, Shasha Lu)
  • rgw: don’t hold data_lock over frontend io (pr#20621, Casey Bodley)
  • rgw: don’t leak S3 LDAPHelper (pr#12427, Matt Benjamin)
  • rgw: dont log EBUSY errors in ‘sync error list’ (issue#22473, pr#19580, Casey Bodley)
  • rgw: dont reuse stale RGWObjectCtx for get_bucket_info() (issue#21506, pr#17916, Casey Bodley)
  • rgw: don’t write bucket_header when it is not changed in bucket_link/unlink (pr#17356, Shasha Lu)
  • rgw: don’t write bucket_header when it is not changed in rgw_bucket_prepare_op (pr#18763, Xinying Song)
  • rgw: download object might fail for local invariable uninitialized (issue#23146, pr#20612, fang yuxiang)
  • rgw: drop a repeated statement for encode_xml() (pr#20195, luomuyao)
  • rgw: drop commented functions (pr#19671, Jos Collin)
  • rgw: drop dump_uri_from_state() which isn’t used anymore (pr#19924, Radoslaw Zarzynski)
  • rgw: drop iter in rgw_op.cc (pr#19583, Bingyin Zhang)
  • rgw: drop marker in RGWLC::process() (pr#19591, Bingyin Zhang)
  • rgw: drop outdated function doc (pr#18370, Jiaying Ren)
  • rgw: drop “realm remove” in radosgw-admin (pr#18212, Shasha Lu)
  • rgw: drop redundant RGW_OP_STAT_OBJ check (pr#19933, Bingyin Zhang)
  • rgw: drop the unnecessary handling of Swift’s X-Storage-Policy on objects (pr#16383, Jiaying Ren)
  • rgw: drop the unused function init_anon_user() (pr#16874, Radoslaw Zarzynski)
  • rgw: Drop unnecessary return (pr#17520, Jos Collin)
  • rgw: drop unused function apply_epoch (pr#17593, Shasha Lu)
  • rgw: drop unused iter in XMLObj::find_first (pr#19709, luomuyao)
  • rgw: drop unused variable bucket_instance_ids (pr#19708, Bingyin Zhang)
  • rgw: drop unused variable in copy_obj_data() (pr#18477, Enming Zhang)
  • rgw: drop unused vector elements (pr#19815, Bingyin Zhang)
  • rgw: drop useless includes in rgw_{main.cc, common.h} (pr#19109, Jiaying Ren)
  • rgw: drop useless lines (pr#19817, Bingyin Zhang)
  • rgw: drop useless type conversion (pr#19824, Bingyin Zhang)
  • rgw: drop variable bl in rgw_op.cc (pr#19584, Bingyin Zhang)
  • rgw: Drop #warning TODO (issue#19851, pr#17012, Jos Collin)
  • rgw: dump Last-Modified in Swift’s responses for GET/HEAD on container (issue#20883, pr#16757, Radoslaw Zarzynski)
  • rgw: enable ‘qlen’ & ‘qactive’ performance counters (pr#20842, Mark Kogan)
  • rgw: encoding fixes (issue#23779, pr#21500, Yehuda Sadeh)
  • rgw: Error check on return of read_line() (pr#17880, Amit Kumar)
  • rgw: es module: set compression type correctly (issue#22758, pr#20796, Abhishek Lekshmanan)
  • rgw: evaluate the correct bucket action for GetACL bucket operation (issue#21013, pr#17050, Abhishek Lekshmanan)
  • rgw: exit early if rgw_bucket_set_attrs() fails (pr#17041, dengxiafubi)
    rgw: Expire entries in bucket info cache
  • rgw_file: fix write error when the write offset overlaps (issue#21455, pr#17809, Yao Zongyou)
  • rgw: fix a bug in rgw cache in delete_system_obj and get_system_obj (pr#10992, zhangshaowen)
  • rgw: fix accessing expired memory in PrefixableSignatureHelper (issue#21085, pr#17206, Radoslaw Zarzynski)
  • rgw: fix a typo in comment (pr#19608, luomuyao)
  • rgw: fix a typo in comment (pr#20164, luomuyao)
  • rgw: fix a typo in comment (pr#20355, luomuyao)
  • rgw: fix a typo in rgw_perms[] (pr#20024, luomuyao)
  • rgw: fix bilog entries on multipart complete (issue#21772, pr#18271, Casey Bodley)
  • rgw: fix BZ 1500904, stale bucket index entry remains after obj delete (pr#18709, J. Eric Ivancich)
  • rgw: fix chained cache invalidation to prevent cache size growth (issue#22410, pr#19455, Mark Kogan)
  • rgw: Fix closing tag for Prefix (pr#17663, Shasha Lu)
  • rgw: fix cls_bucket_head result order consistency (pr#18700, Tianshan Qu)
  • rgw: fix collect()’s return in coroutine (pr#19606, Xinying Song)
  • rgw: fix command argument error for radosgw-admin (issue#21723, pr#18175, Yao Zongyou)
  • rgw: fix ‘copy part’ without ‘x-amz-copy-source-range’ (issue#22729, pr#20002, Malcolm Lee)
  • rgw: fix ‘copy part’ without ‘x-amz-copy-source-range’ when compression enabled (issue#23196, pr#20686, fang yuxiang)
  • rgw: fix crash with rgw_run_sync_thread false (issue#20448, pr#20769, Orit Wasserman)
  • rgw: Fix dereference of empty optional (issue#21962, pr#18602, Adam C. Emerson)
  • rgw: fix error handling for GET with ?torrent (issue#23506, pr#21576, Casey Bodley)
  • rgw: fix error handling in Browser Uploads (pr#15054, Radoslaw Zarzynski)
  • rgw: fix error handling in ListBucketIndexesCR (issue#21735, pr#18198, Casey Bodley)
  • rgw: fixes for multisite replication of encrypted objects (issue#20668, issue#20671, pr#16612, Casey Bodley)
  • rgw: fix extra_data_len handling in PutObj filters (issue#21895, pr#18489, Casey Bodley)
  • rgw: fix for empty query string in beast frontend (issue#22797, pr#20120, Casey Bodley)
  • rgw: fix for issue #21647 (issue#23859, pr#21647, Yehuda Sadeh)
  • rgw: fix for pause in beast frontend (issue#21831, pr#18402, Casey Bodley)
  • rgw: fix for usage truncated flag (pr#20926, Yehuda Sadeh, Greg Farnum, Robin H. Johnson)
  • rgw: Fix getter function names in RGWEnv (pr#18377, Jos Collin)
  • rgw: fix GET website response error code (issue#22272, pr#19236, Dmitry Plyakin)
  • rgw: fix handling of ENOENT in RGWRadosGetOmapKeysCR (pr#19878, Casey Bodley)
  • rgw: fix index cancel op miss update header (pr#20396, Tianshan Qu)
  • rgw: Fix infinite call for bi list when resharding a bucket (issue#22721, pr#21584, Orit Wasserman)
  • rgw: fix lc process only schdule the first item of lc objects (issue#21022, pr#17061, Shasha Lu)
  • rgw:fix list objects with marker wrong result when bucket is enable versioning (issue#21500, pr#17934, yuliyang)
  • rgw: fix memory fragmentation problem reading data from client (pr#20724, Marcus Watts)
  • rgw: Fix multisite Synchronization failed when read and write delete … (issue#22804, pr#20814, Niu Pengju)
  • rgw: fix not responding when receiving SIGHUP signal (pr#16854, Yao Zongyou)
  • rgw: fix null pointer crush (pr#18861, Sibei Gao)
  • rgw: fix obj copied from remote gateway acl full_control issue (issue#20658, pr#16127, Enming Zhang)
  • rgw: fix opslog cannot record remote_addr (issue#20931, pr#16860, Jiaying Ren)
  • rgw: fix opslog can’t record referrer when using curl as client (issue#20935, pr#16863, Jiaying Ren)
  • rgw: fix opslog uri as per Amazon s3 (issue#20971, pr#16958, Jiaying Ren)
  • rgw: fix radosgw-admin bucket rm with –purge-objects and –bypass-gc (issue#22122, issue#19959, pr#18922, Aleksei Gutikov)
  • rgw: fix radosgw-admin quota enable return value bug (issue#21608, pr#18057, baixueyu)
  • rgw: fix radosgw linkage with WITH_RADOSGW_BEAST_FRONTEND=OFF (issue#23680, pr#21380, Casey Bodley)
  • rgw: fix recursive lock (pr#19430, Tianshan Qu)
  • rgw: fix resource leak in rgw_bucket.cc and rgw_user.cc (issue#21214, pr#17353, Luo Kexue)
  • rgw: fix return value of auth v2/v4 (issue#22439, pr#19310, Bingyin Zhang)
  • rgw: fix rewrite a versioning object create a new object bug (issue#21984, pr#18662, Enming Zhang)
  • rgw: fix rewrite options usage text (pr#18968, Jos Collin)
  • rgw: fix RGWCompletionManager get_next stuck after going down (issue#22799, pr#20095, Tianshan Qu)
  • rgw: fix RGWLibIO did not init RGWEnv (pr#19065, Tianshan Qu)
  • rgw: fix s3 website redirection error (pr#19252, yuliyang)
  • rgw: fix s3website redirect location string length (pr#19826, yuliyang)
  • rgw: fix Swift container naming rules (issue#19264, pr#13992, Robin H. Johnson)
  • rgw: Fix swift object expiry not deleting objects (issue#22084, pr#18821, Pavan Rallabhandi)
  • rgw: fix sync status conflict with cloud sync (pr#21425, Casey Bodley)
  • rgw: fix the bug of radowgw-admin zonegroup set requires realm (issue#21583, pr#19061, lvshanchun)
  • rgw: fix the max-uploads parameter not work (issue#22825, pr#20158, Xin Liao)
  • rgw: fix the return type is wrong (pr#19773, hechuang)
  • rgw: fix total_time to msec as per AWS S3 (pr#17541, Jiaying Ren)
  • rgw: fix typo anynoymous to anonymous (pr#19281, Bingyin Zhang)
  • rgw: fix typo compete to complete (pr#19675, Bingyin Zhang)
  • rgw: Fix typo in comment (pr#21032, Simran Singhal)
  • rgw: fix typo in GetOmapKeysCR (pr#19713, lvshanchun)
  • rgw: fix typo signle to single (pr#19517, Bingyin Zhang)
  • rgw: fix typo woild to would (pr#19472, Bingyin Zhang)
  • rgw: Fix use after free in IAM policy parser (pr#16823, Adam C. Emerson)
  • rgw: fix use of libcurl with empty header values (issue#23663, pr#21358, Casey Bodley)
  • rgw: format logs in file rgw_lc.cc (pr#19615, Bingyin Zhang)
  • rgw: format rgw_bucket_dir_header in ceph-dencoder (pr#19753, Bingyin Zhang)
  • rgw: gc use aio (pr#20546, Yehuda Sadeh)
    rgw: Handle stale bucket info in RGWDeleteBucketPolicy rgw: Handle stale bucket info in RGWDeleteBucketWebsite rgw: Handle stale bucket info in RGWPutBucketPolicy rgw: Handle stale bucket info in RGWPutMetadataBucket rgw: Handle stale bucket info in RGWSetBucketVersioning rgw: Handle stale bucket info in RGWSetBucketWebsite
  • rgw: honor the tenant part of rgw_bucket during comparisons (issue#20897, pr#16796, Radoslaw Zarzynski)
  • rgw: iam policy printing cleanups (pr#18961, Kefu Chai)
  • rgw: Ignoring the returned error (pr#17907, Amit Kumar)
  • rgw: implement ipv4 aws:SourceIp condition for bucket policy (pr#19167, yuliyang)
  • rgw: improve handling of Swift’s error messages and limits (issue#17938, issue#21169, issue#17935, issue#17934, issue#17936, pr#15369, Radoslaw Zarzynski)
  • rgw: improve sync status: display behind bucket shards (pr#20027, lvshanchun)
  • rgw: improve sync status (pr#19573, lvshanchun)
  • rgw: include SSE-KMS headers in encrypted upload response (issue#21576, pr#17999, Casey Bodley)
  • rgw: incorporate the Transfer-Encoding fix for CivetWeb (issue#21015, pr#17072, Radoslaw Zarzynski)
  • rgw: Initialization of epoch,len (pr#17722, Amit Kumar)
  • rgw: Initialize is_master, max_aio, size (pr#16933, amitkuma)
  • rgw: Initializes uninitialized members (pr#16855, Amit Kumar)
  • rgw: init oldest period after setting run_sync_thread (issue#21996, pr#18664, Orit Wasserman, Casey Bodley)
  • rgw: keep compression type consistent between parts of s3 Multipart (pr#19740, fang yuxiang)
  • rgw: keystone: bump up logging when error is received (issue#22151, pr#18985, Abhishek Lekshmanan)
  • rgw:lc fix expiration time (issue#21533, pr#17824, Shasha Lu)
  • rgw: lc support Content-MD5 request header and fix a rgw crash bug (issue#21980, pr#18534, Enming Zhang)
  • rgw: lease_cr->go_down is called twice, remove the needless one (pr#19394, Zhang Shaowen)
  • rgw: librgw: export multitenancy support (pr#19358, Tao Chen)
  • rgw: librgw: fix shutdown err with resources uncleaned (issue#22296, pr#19279, Tao Chen)
  • rgw: lifecycle omap entry was removed in abnormal situation (pr#19921, fang yuxiang)
  • rgw: list_objects() honors end_marker regardless of namespace (issue#18977, pr#15273, Radoslaw Zarzynski)
  • rgw: loadgen fix generate random object name rgw crash issue (issue#22006, pr#18536, Enming Zhang)
  • rgw: log the right http status code in civetweb frontend’s access log (issue#22538, pr#19678, Yao Zongyou)
  • rgw: log unlink_instance mtime as object’s mtime (issue#18885, pr#20016, Yehuda Sadeh)
  • rgw: lttng: Trace rgw data transfer, bi entry and object header update processes (pr#20556, Yang Honggang)
  • rgw: make init env methods return an error (issue#23039, pr#20488, Abhishek Lekshmanan)
  • rgw: make radosgw object stat RGW_ATTR_COMPRESSION dump readable (pr#19846, fang yuxiang)
  • rgw: mfa support (pr#19283, Yehuda Sadeh)
  • rgw: mimic: rgw: policy: modify s3:ListBucketMultiPartUploads to s3:ListBucketMul (issue#24062, pr#21916, xiangxiang)
  • rgw: modify s3 type subuser access permissions fail through admin rest api (issue#21983, pr#18641, yuliyang)
  • rgw: move all pool creation into rgw_init_ioctx (issue#23480, pr#21534, Casey Bodley)
  • rgw: mrgw.sh uses instance name ‘client.rgw’ (pr#18404, Casey Bodley)
  • rgw: multisite log tracing (pr#16492, Yehuda Sadeh, Casey Bodley)
  • rgw,nfs: Add hint to use -o sync when mouting (pr#16210, Adam Kupczyk)
  • rgw: no need to deal with md5 header in get_data (pr#19144, Zhang Shaowen)
  • rgw: optimize function abort_bucket_multiparts (pr#19710, Bingyin Zhang)
  • rgw: optimize function bucket_lc_prepare (pr#19613, Bingyin Zhang)
  • rgw: optimize function parse_raw_oid (pr#19814, Bingyin Zhang)
  • rgw: optimize function RGWHandler::do_init_permissions (pr#19700, Bingyin Zhang)
  • rgw: optimize memory usage in function rgw_bucket::get_key (pr#19391, Bingyin Zhang)
  • rgw: optimize next start time for lifecycle (pr#19596, Bingyin Zhang)
  • rgw: optimize the rgw-attr del code logic (pr#18895, wangsongbo)
  • rgw: optimize time skew check (pr#19511, Bingyin Zhang)
  • rgw: parse old rgw_obj with namespace correctly (issue#22982, pr#20425, Yehuda Sadeh)
  • rgw: policy: support for s3 conditionals in ListBucket Op (pr#16628, Abhishek Lekshmanan)
  • rgw: Potential fix for possible 500 on POST (pr#18954, Adam C. Emerson)
  • rgw: Prevent overflow of cached stats values (issue#20934, pr#17116, Aleksei Gutikov)
  • rgw: proper error message when tier_type does not exist (issue#22469, pr#19575, lvshanchun, Chang Liu)
  • rgw: pull up beast submodule and update frontend (pr#17923, Casey Bodley)
  • rgw: put bucket policy panics RGW process (issue#22541, pr#19687, Bingyin Zhang)
  • rgw: radosgw-admin abort early for user stats for empty uids (issue#23322, pr#20846, Abhishek Lekshmanan)
  • rgw: radosgw-admin should not use metadata cache for readonly commands (issue#23468, pr#21129, Orit Wasserman)
  • rgw: radosgw-admin zonegroup get and zone get return defaults when there is no realm (issue#21615, pr#18667, lvshanchun)
  • rgw: radosgw: fix awsv4 header line sort order (issue#21607, pr#18046, Marcus Watts)
  • rgw: radosgw: usage: fix bytes_sent bug (issue#19870, pr#16834, Marcus Watts)
  • rgw: raise log level on coroutine shutdown errors (issue#23974, pr#21791, Casey Bodley)
  • rgw: Reinstating error codes mapping for Roles (pr#20309, Pritha Srivastava)
  • rgw: reject encrypted object COPY before supported (issue#23232, pr#20739, Jeegn Chen)
  • rgw: release cls lock if taken in RGWCompleteMultipart (issue#21596, pr#18104, Matt Benjamin)
  • rgw: Remove assertions in IAM Policy (pr#18225, Adam C. Emerson)
  • rgw: remove get_system_obj_attrs in function RGWDeleteLC::execute and RGWDeleteCORS::execute (pr#19582, Bingyin Zhang)
  • rgw: remove placement_rule from rgw_link_bucket() (issue#21990, pr#18657, Casey Bodley)
  • rgw: remove redundant parenthesis in logs (pr#19375, Bingyin Zhang)
  • rgw: remove redundant S3AnonymousEngine (pr#19474, Bingyin Zhang)
  • rgw: remove redundant signature compare in LocalEngine::authenticate (pr#19676, Bingyin Zhang)
  • rgw: Remove the useless output when list zones (pr#17434, iliul)
  • rgw: remove unused cls_user_add_bucket (pr#19917, Yao Zongyou)
  • rgw: remove unused disable_signal_fd (pr#18875, Yao Zongyou)
  • rgw: remove unused function get_system_obj_attrs (pr#19852, Yao Zongyou)
  • rgw: Remove unused Parameter in Function RGWConf::init() (pr#17129, Wen Zhang)
  • rgw: remove unused param in AWSGeneralAbstractor::get_auth_data_v4 (pr#19250, Bingyin Zhang)
  • rgw: remove unused param in get_bucket_instance_policy_from_attr (pr#19129, Bingyin Zhang)
  • rgw: remove unused variables (pr#16649, Zhang Lei)
  • rgw: remove useless lines in RGWDeleteBucket::execute (pr#19699, Bingyin Zhang)
  • rgw: reshard cancel command should clear bucket resharding flag (issue#21619, pr#21120, Orit Wasserman)
  • rgw: reshard should not update stats when linking new bucket instance (issue#22124, pr#19253, Orit Wasserman)
  • rgw: retry CORS put/delete operations on ECANCELLED (issue#22517, pr#19601, Adam C. Emerson)
  • rgw: return ‘Access-Control-Allow-Origin’ header when the set and delete bucket website through XMLHttpRequest (pr#17632, yuliyang)
  • rgw: return ‘Access-Control-Allow-Origin’ header when the set bucket versioning through XMLHttpRequest (pr#17631, yuliyang)
  • rgw: return bucket’s location no matter which zonegroup it located in (issue#21125, pr#17250, Shasha Lu)
  • rgw: return EINVAL if max_keys can not convert correctly (issue#23586, pr#21285, yuliyang)
  • rgw: Return Error if Bucket Policy Contians Undefined Action (pr#17433, zhangwen)
  • rgw: Returning when dst_ioctx.operate() returns error (pr#17873, Amit Kumar)
  • rgw: return valid Location element, CompleteMultipartUpload (pr#19902, Matt Benjamin)
  • rgw: revert PR #8765 (pr#16807, fang.yuxiang)
  • rgw: Revert “radosgw: fix awsv4 header line sort order.” (issue#21832, pr#18381, Casey Bodley)
  • rgw: Revert “rgw_file: disable FLAG_EXACT_MATCH enforcement” (issue#22827, pr#20171, Matt Benjamin)
  • rgw: Revert “rgw: reshard should not update stats when linking new bucket instance” (pr#20052, Orit Wasserman)
  • rgw: rework json/xml escape usage follow #19806 (pr#19845, fang yuxiang)
  • rgw: rgw-admin: check input parameters for friendly prompt (pr#17343, Yao Zongyou)
  • rgw: rgw-admin: check the data extra pool supports omap (pr#18978, Yao Zongyou)
  • rgw: rgw-admin: properly filtering bucket stats by user_id or bucket_name (pr#19401, Yao Zongyou)
  • rgw: rgw-admin: require –yes-i-really-mean-it when using –inconsistent_index (issue#20777, pr#17185, Orit Wasserman)
  • rgw: rgw-admin: support for processing all gc objects including unexpired (pr#17482, Yao Zongyou)
  • rgw: RGW: change function parameters from value to refrence (pr#18355, Sibei Gao)
  • rgw: RGWCivetWeb::read_data: fix arguments to mg_read() call (issue#23596, pr#21291, Nathan Cutler)
  • rgw: rgw clean-up: remove unreferenced pure virtual class StreamObjData (pr#18799, J. Eric Ivancich)
  • rgw: rgw clean-up: remove unused var & func in RGWRados::SystemObject (pr#18987, J. Eric Ivancich)
  • rgw: rgw cleanup: some unnecessary function called and repeated assignment (pr#18817, Enming Zhang)
  • rgw: rgw cloud sync (issue#21802, pr#18932, lvshanchun, Yehuda Sadeh, Chang Liu, Abhishek Lekshmanan)
  • rgw: RGWEnv::set() takes std::string (issue#22101, pr#18866, Casey Bodley)
  • rgw: rgw_file: alternate fix deadlock on lru eviction (pr#20034, Matt Benjamin)
  • rgw: rgw_file: avoid evaluating nullptr for readdir offset (pr#20145, Matt Benjamin)
  • rgw: rgw_file: conditionally unlink handles when direct deleted (issue#23299, pr#20834, Matt Benjamin)
  • rgw: rgw_file: explicit NFSv3 open() emulation (pr#18365, Matt Benjamin)
  • rgw: rgw_file: fix LRU lane lock in evict_block() (issue#21141, pr#17267, Matt Benjamin)
  • rgw: rgw_file: implement variant offset readdir processing (pr#18335, Matt Benjamin)
  • rgw: rgw_file: introduce new fsid and rgw_mount (pr#15330, Gui Hecheng)
  • rgw: rgw_file: set s->obj_size from bytes_written (issue#21940, pr#18571, Matt Benjamin)
  • rgw: rgw_file: Silence unused-function warnings (pr#19278, Brad Hubbard)
  • rgw: RGW: fix a bug about inconsistent unit of comparison (issue#21590, pr#17958, gaosibei)
  • rgw: rgw.iam: change ‘1’ to ‘1ULL’ in function print_actions (pr#18900, Bingyin Zhang)
  • rgw: rgw_lc: add support for optional filter argument and make ID optional (issue#19587, issue#20872, pr#16818, Abhishek Lekshmanan)
  • rgw: rgw_lc: support for AWSv4 authentication (pr#16734, Abhishek Lekshmanan)
  • rgw: rgw_log, rgw_file: account for new required envvars (issue#21942, pr#18572, Matt Benjamin)
  • rgw: Rgw master fix plus (issue#21000, issue#21003, issue#20501, pr#17040, Zhang Shaowen, Marcus Watts)
  • rgw: rgw, mon: normalize delete/remove in admin console (cleanup 22444) (issue#14363, issue#22444, pr#19439, Jesse Williamson)
  • rgw: RGW: Multipart upload may double the quota (issue#21586, pr#17959, Sibei Gao)
  • rgw: rgw multisite: automated trimming for bucket index logs (issue#18229, pr#17761, Casey Bodley)
  • rgw: RGW NFS: mount cmdline example missing -osync (pr#15855, Matt Benjamin)
  • rgw: RGW-NFS: Use rados cluster_stat to report filesystem usage (issue#22202, pr#20093, Supriti Singh)
  • rgw: rgw_op: Drop the Old LifecycleConfiguration from logs (pr#16821, Abhishek Lekshmanan)
  • rgw: rgw_op: exit early if object has no attrs in GetObjectTagging (issue#21010, pr#17048, Abhishek Lekshmanan)
  • rgw: RGWPutLC return ERR_MALFORMED_XML when missing <Rule> tag in lifecycle.xml (issue#21377, pr#17683, Shasha Lu)
  • rgw: rgw_put_system_obj takes bufferlist (pr#19897, Casey Bodley)
  • rgw: rgw_rados: set_attrs now sets the same time for BI & object (issue#21200, pr#17400, Abhishek Lekshmanan)
  • rgw: rgw/rgw_op.cc: Fix error message in rgw_user_get_all_buckets_stats (pr#18781, iliul)
  • rgw: rgw: source data in ‘default.rgw.buckets.data’ may not be deleted after inter-bucket copy (issue#21819, pr#18369, baixueyu)
  • rgw: RGW: support for tagging in lifecycle policies (pr#17305, Abhishek Lekshmanan)
  • rgw: RGW: update S3 POST policy handling of Content-Type (issue#20201, pr#18658, Matt Benjamin)
  • rgw: rgw: use camelcase format in request headers (pr#19210, lvshanchun, Chang Liu)
  • rgw: RGWUser::init no longer overwrites user_id (issue#21685, pr#18137, Casey Bodley)
  • rgw: S3 Bucket Policy Conditions IpAddress and NotIpAddress do not work (issue#20991, pr#17010, John Gibson)
  • rgw: s3website error handler uses original object name (issue#23201, pr#20693, Casey Bodley)
  • rgw:send x-amz-version-id header when upload files (pr#18935, Xinying Song)
  • rgw: set bucket versioninig donot change versioning status if missing status in xml (issue#21364, pr#17662, Shasha Lu)
  • rgw: set num_shards on ‘radosgw-admin data sync init’ (issue#22083, pr#18883, Casey Bodley)
  • rgw: set priority on perf counters (pr#20006, John Spray)
  • rgw: set sync_from_all as true when no value is seen (issue#22062, pr#18926, Abhishek Lekshmanan)
  • rgw: setup locks for libopenssl (issue#22951, issue#23203, pr#20390, Abhishek Lekshmanan, Jesse Williamson)
  • rgw: share time skew check between v2 and v4 auth (pr#20013, Casey Bodley)
  • rgw: Silence maybe-uninitialized false positives (pr#19274, Brad Hubbard)
  • rgw: silence not allow register storage class specifier warning (pr#19859, Yao Zongyou)
  • rgw: simplify use of map::emplace in iam (pr#18706, Casey Bodley)
  • rgw: Small refactor and two bug fixes (issue#21901, issue#21896, pr#18606, Adam C. Emerson)
  • rgw: some cleanup for sync status (pr#20894, Enming Zhang)
  • rgw: stop/join TokenCache revoke thread only if started (issue#21666, pr#18106, Karol Mroz)
  • rgw: stream metadata full sync init (issue#18079, pr#12429, Yehuda Sadeh)
  • rgw: submodule: update Beast to ceph/ceph-master branch (pr#19182, Casey Bodley)
  • rgw: switch beast frontend back to stackful coroutine (issue#20048, pr#20449, Casey Bodley)
  • rgw: sync tracing fixes (issue#22833, pr#20191, Yehuda Sadeh)
  • rgw: tenant fixes for dynamic resharding (issue#22046, pr#18811, Orit Wasserman)
  • rgw,tests: fix s3atests that are failing for sometime (pr#20678, Vasu Kulkarni)
  • rgw,tests: qa: fix overrides for openssl_keys task (pr#20981, Casey Bodley)
  • rgw,tests: qa: re enable LC tests (pr#17020, Abhishek Lekshmanan)
  • rgw,tests: qa/rgw: add beast frontend to some rgw suites (pr#17977, Casey Bodley)
  • rgw,tests: qa/rgw: combine swift, s3tests, ragweed into single verify task (pr#20756, Casey Bodley)
  • rgw,tests: qa/rgw: disable log trim in multisite suite (pr#19438, Casey Bodley)
  • rgw,tests: qa/rgw: hadoop-s3a suite targets centos_latest (pr#17777, Casey Bodley)
  • rgw,tests: qa/rgw: ignore errors from ‘pool application enable’ (issue#21715, pr#18193, Casey Bodley)
  • rgw,tests: qa/rgw: remove some civetweb overrides for beast testing (issue#23002, pr#20440, Casey Bodley)
  • rgw,tests: qa/rgw: renamed ssl task to openssl_keys (pr#20863, Ricardo Dias)
  • rgw,tests: qa/rgw: use ‘ceph osd pool application enable’ on created pools (pr#17162, Casey Bodley)
  • rgw,tests: qa/rgw: verify suite tests civetweb with ssl (pr#20444, Casey Bodley)
  • rgw,tests: qa/smoke: add rgw crypto config for s3tests (pr#17700, Casey Bodley)
  • rgw,tests: qa/tasks/swift: add support for the “force-branch” configurable (pr#21027, Radoslaw Zarzynski)
  • rgw,tests: rgw, qa: integrate Tempest to verify RadosGW’s compliance with Swift API (pr#16344, Radoslaw Zarzynski)
  • rgw,tests: test/rgw: fix test_encrypted_object_sync for 3+ zones (pr#17377, Casey Bodley)
  • rgw: the metavariables in frontends-related config won’t be expanded (pr#19689, root)
  • rgw,tools: tools/rgw: add script to inspect admin socket “cr dump” (pr#15554, Casey Bodley)
  • rgw: Torrents are not supported for objects encrypted using SSE-C (issue#21720, pr#17956, Zhang Shaowen)
  • rgw: trim all spaces inside a metadata value (issue#23301, pr#20841, Orit Wasserman)
  • rgw: udpate radosgw-admin usage with bi purge (pr#18245, Yao Zongyou)
  • rgw: unlink deleted bucket from bucket’s owner (issue#22248, pr#20017, Casey Bodley)
  • rgw: unreachable return in RGWRados::trim_bi_log_entries (pr#17367, Amit Kumar)
  • rgw: update life cycle related log level (pr#18845, Yao Zongyou)
  • rgw: update outdated debug func name (pr#17440, Jiaying Ren)
  • rgw: update quota is inconsistent at add/del object with compression (issue#22568, pr#19772, fang yuxiang)
  • rgw: update the usage read iterator in truncated scenario (issue#21196, pr#17939, Mark Kogan)
  • rgw: update usage() with status (pr#18178, Jos Collin)
  • rgw: update vstart.sh to support rgw ssl port notation : ‘–rgw_port 443s’ (issue#21151, pr#17989, Mark Kogan)
  • rgw: upldate the max-buckets when the quota is uploaded (pr#20063, zhaokun)
  • rgw: URL-decode S3 and Swift object-copy URLs (issue#22121, pr#19936, Matt Benjamin)
  • rgw: url_encode key name and instance in es sync module (pr#20707, Chang Liu)
  • rgw: use explicit index pool placement (issue#22928, pr#20352, Yehuda Sadeh)
  • rgw: Use namespace for lc_pool and roles_pool (issue#20177, pr#16889, Orit Wasserman)
  • rgw: Various cleanups and options update in rgw_admin.cc (pr#18302, Jos Collin)
  • rgw: vstart.sh: fix mstop.sh can not stop rgw (pr#17438, Jiaying Ren)
  • rgw: ‘zone placement’ commands validate compression type (issue#21775, pr#18273, Casey Bodley)
  • rocksdb: sync with upstream (issue#21603, pr#18262, Kefu Chai)
  • rpm: rm macros in comments (issue#22250, pr#17070, Ken Dreyer)
  • script/build-integration-branch: check errors (pr#17578, Sage Weil)
  • script/build-integration-branch: python3 compatible and pep8 clean (pr#18035, Kefu Chai)
  • scripts: new backport-create-issue script (pr#21480, Nathan Cutler)
  • selinux: Allow ceph to execute ldconfig (pr#21974, Boris Ranto)
  • selinux: Allow getattr on lnk sysfs files (pr#17891, Boris Ranto)
  • spdk: advance to upstream dc82989d (pr#20713, Nathan Cutler)
  • src: fix various log messages (pr#21112, Gu Zhongyan)
  • src/msg/rdma: fixes failure on assert in notify() (pr#17007, Alex Mikheev)
  • suites/cephmetrics: Add Centos 7 (pr#18594, Zack Cerza)
  • test: assert check for negative returns (pr#17296, Amit Kumar)
  • test/fio: generate db histogram to help debug rocksdb performance (pr#16808, Pan Liu, Xiaoyan Li)
  • test: fix bash path in shebangs (part 2) (pr#17955, Alan Somers)
  • test: fix CLI unit formatting tests (pr#22260, Jason Dillaman)
  • test: Incorrect conversion to double (pr#18963, Amit Kumar)
  • test/librados: reorder ASSERT_EQ() arguments (pr#16625, Yan Jun)
  • test,osd,kvstore_tool: silence warnings and prepare test buffer in the right way (pr#18406, Adam C. Emerson)
  • tests: bluestore/fio: Fixed problem with all objects having the same hash (pr#17770, Adam Kupczyk)
  • tests: CentOS 7.4 is now the latest (pr#17776, Nathan Cutler)
  • tests - ceph-ansible vars additions (issue#21822, pr#18378, Yuri Weinstein)
  • tests: ceph-disk: ignore E722 in flake8 test (issue#22207, pr#19072, Nathan Cutler)
  • tests: ceph-disk: mock get fsid (pr#19254, Kefu Chai)
  • tests: ceph-disk: Remove sitepackages=True (issue#22823, pr#20151, Brad Hubbard)
  • tests: ceph-objectstore-tool: don’t destroy SnapMapper until the txn is completed (issue#23121, pr#20593, Kefu Chai)
  • tests: Changes required for teuthology’s systemd support (pr#18380, Zack Cerza)
  • tests: Check for empty output in test_dump_pgstate_history (pr#20838, Brad Hubbard)
  • tests: cleanup: drop calamari tasks (pr#17531, Nathan Cutler)
  • tests: cleanup: drop upgrade/jewel-x/point-to-point-x (issue#22888, pr#20245, Nathan Cutler)
  • tests: cmake,test/mgr: restructure dashboard tests and cmake related fixes (pr#20768, Kefu Chai)
  • tests: common/obj_bencher: set {min,max}_iops if runtime < 1 sec (pr#17182, Kefu Chai)
  • tests: c_read_operations.cc: Silence tautological-compare compiler warning (pr#19953, Brad Hubbard)
  • tests: fix uninitialized value found by coverity scan (pr#17895, J. Eric Ivancich)
  • tests: Increase sleep in test_pidfile.sh (pr#17052, David Zafman)
  • tests: librgw_file: remove unused using statement (pr#17085, Yao Zongyou)
  • tests: mark_unfound_lost fix and some other minor changes (issue#21907, pr#18449, David Zafman)
  • tests: mgr/dashboard: Allow sourcing run-backend-api-tests.sh (pr#20874, Sebastian Wagner)
  • tests: mgr/dashboard: create venv for running tox (pr#21490, Kefu Chai)
  • tests: mgr/dashboard: notification queue: fix priority tests (pr#21147, Ricardo Dias)
  • tests: mimic: qa: fix test on “ceph fs set cephfs allow_new_snaps” (pr#21830, Kefu Chai)
  • tests: mimic: qa/workunits/rados/test_envlibrados_for_rocksdb: install g++ not g++-4.7 (pr#22117, Kefu Chai)
  • tests: mimic: test: Need to escape parens in log-whitelist for grep (pr#22075, David Zafman)
  • tests: mimic: test: wait_for_pg_stats() should do another check after last 13 secon… (pr#22199, David Zafman)
  • tests: misc: Fix bash path in shebangs (pr#16494, Alan Somers)
  • tests: mstart.sh: remove bashizm in /bin/sh script (pr#18541, Mykola Golub)
  • tests: point-to-point-x: upgrade client.1 to -x along with cluster nodes (issue#21499, pr#17910, Nathan Cutler)
  • tests: qa: add cbt task for performance testing (pr#17583, Neha Ojha)
  • tests: qa: add cosbench workloads and override teuthology default settings (pr#21710, Neha Ojha)
  • tests/qa: Adding $ distro mix - rgw (pr#22070, Yuri Weinstein)
  • tests/qa: adding rados/.. dirs (pr#22068, Yuri Weinstein)
  • tests: qa: add “restful” to ceph_mgr_modules in ceph-ansible suite (pr#18634, Kefu Chai)
  • tests: qa: add simple and dirty script to find ports being used (pr#19102, Joao Eduardo Luis)
  • tests: qa: big: add openstack.yaml (pr#16864, Nathan Cutler)
  • tests: qa: clean up dnsmasq task and fix EPERM error (pr#20680, Casey Bodley)
  • tests: qa: create_cache_pool no longer runs ‘pool application enable’ (issue#21155, pr#17312, Casey Bodley)
  • tests: qa: decrease the msg_inject_socket_failures from 1/500 to 1/1000 (issue#22093, pr#19542, Kefu Chai)
  • tests: qa: disable mon-health-to-clog in upgrade test (pr#19233, Kefu Chai)
  • tests: qa: disable -Werror when compiling env_librados_test (pr#21433, Kefu Chai)
  • tests: qa: do not “ceph fs get cephfs” w/o a cephfs fs (pr#18533, Kefu Chai)
  • tests: qa: do not wait for down/out osd for pg convergence (pr#18808, Kefu Chai)
  • tests/qa - enabled ceph-deploy runs on mira nodes (pr#21253, Yuri Weinstein)
  • tests: qa: fix pool-quota related tests (issue#21409, pr#17763, xie xingguo)
  • tests: qa: Fix shebangs on openstack scripts (pr#16546, Alan Somers)
  • tests: qa: reduce “mon client hunt interval max multiple” to 2 for all clients (pr#21658, Kefu Chai)
  • tests: qa: reduce mon-client-hunt-interval-max-multiple to 2 (pr#18283, Kefu Chai)
  • tests: qa: revert “qa: use config_path property instead of literal” (pr#17850, Patrick Donnelly)
  • tests: qa/run-standalone.sh: set PYTHONPATH for FreeBSD also (pr#20646, Kefu Chai)
  • tests: qa: s/backfill/backfilling/ (pr#18235, Kefu Chai)
  • tests: qa/stanalone: pass options using –<option-name>=<value> (pr#19544, Kefu Chai)
  • tests: qa/standalone: Add trap for signals to restore the kernel core pattern (pr#17026, David Zafman)
  • tests: qa/standalone/ceph-helpers.sh: provide argument to dirname (issue#23805, pr#21552, Nathan Cutler)
  • tests: qa/standalone/ceph-helpers.sh: silence ceph-disk DEPRECATION_WARNING (pr#19478, Kefu Chai)
  • tests: qa/standalone: extract delete_pool() (pr#20634, Kefu Chai)
  • tests: qa/standalone: misc fixes (issue#20465, issue#20921, pr#16709, David Zafman)
  • tests: qa/standalone/mon/misc.sh: Add osdmap-prune tests (issue#23621, pr#21318, Brad Hubbard)
  • tests: qa/standalone/osd/osd-mark-down: create pool to get updated osdmap faster (pr#18191, huangjun)
  • tests: qa/standalone: remove osd-map-max-advance related tests (issue#22596, pr#19816, Kefu Chai)
  • tests: qa/standalone: respect $TEMPDIR (pr#17747, Kefu Chai)
  • tests: qa/standalone/scrub/osd-scrub-repair.sh: add extents flag into object_info_t (issue#21618, pr#18094, xie xingguo)
  • tests: qa/standalone/scrub/osd-scrub-repair.sh: drop omap_digest flag (pr#18150, xie xingguo, Sage Weil)
  • tests: qa/standalone: s/delete_erasure_pool/delete_erasure_coded_pool/ (pr#20667, Kefu Chai)
  • tests: qa: stop testing deprecated “ceph osd create” (issue#21993, pr#18659, Kefu Chai)
  • tests: qa/suites: add minimal performance suite (pr#21104, Neha Ojha)
  • tests: qa/suites/cephmetrics: Updates for new version (pr#21146, Zack Cerza)
  • tests: qa/suites: change fixed-2.yaml users to get 4 openstack disks (pr#16873, Sage Weil)
  • tests: qa/suites: mds.0 -> mds.a (pr#20848, Sage Weil)
  • tests: qa/suites/rados: Disable scrub backoff (issue#23578, pr#21295, Brad Hubbard)
  • tests: qa/suites/rados/mgr/tasks/dashboard: add MDS_ALL_DOWN to whitelist (pr#21549, Ricardo Dias)
  • tests: qa/suites/rados/mgr/tasks/dashboard_v2: add fail_on_skip = false (pr#20925, Ricardo Dias)
  • tests: qa/suites/rados/multimon: whitelist mgr down vs clock skew test (pr#16996, Sage Weil)
  • tests: qa/suites/rados/singleton: more whitelist (pr#19225, Kefu Chai)
  • tests: qa/suites/rados/thrash-old-clients: ms_type=simple (issue#23922, pr#21739, Kefu Chai)
  • tests: qa/suites/rados/upgrade/jewel-x-singleton: tolerate sloppy past_intervals (pr#17293, Kefu Chai)
  • tests: qa/suites/rest/basic/tasks/rest_test: more whitelisting (issue#21425, pr#17794, huangjun)
  • tests: qa/suites/rest/basic/tasks/rest_test: whiltelist OSD_DOWN (issue#21425, pr#18144, huangjun)
  • tests: qa/suites/upgarde/jewel-x/parallel: tolerate mgr warning (pr#17203, Sage Weil)
  • tests: qa/suites/upgarde/jewel-x/point-to-point-x: disable app warnings (pr#16947, Sage Weil)
  • tests: qa/suites: whitelist SLOW_OPS (issue#23495, pr#21324, Kefu Chai)
  • tests: qa/tasks: Add default timeout for wait for pg clean task (pr#21313, Vasu Kulkarni)
  • tests: qa/tasks/ceph_deploy: gatherkeys before mgr deploy (pr#17224, Sage Weil)
  • tests: qa/tasks/ceph_manager: use set_config on revived osd (pr#20901, Neha Ojha)
  • tests: qa/tasks/mgr/dashboard: Fix login expires too soon (pr#21021, Sebastian Wagner)
  • tests: qa/tasks: prolong revive_osd() timeout to 6 min (issue#21474, pr#17902, Kefu Chai)
  • tests: qa/tasks: prolong revive_osd() timeout to 6 min (issue#21474, pr#19024, Kefu Chai)
  • tests: qa/tasks: run cosbench using the CBT task (pr#21656, Neha Ojha)
  • tests: qa/tasks: update ceph-deploy task to use newer ceph-volume syntax (pr#19244, Vasu Kulkarni)
  • tests: qa/tests: Add additional required ceph-ansible vars due to upstream changes (pr#17757, Vasu Kulkarni)
  • tests: qa/tests: add ceph-deploy upgrade tests (issue#20950, pr#16826, Vasu Kulkarni)
  • tests: qa/tests: add openstack volume info + lvs for ceph-volume (pr#20243, Vasu Kulkarni)
  • tests: qa/tests: Fix get_system_type failure due to invalid remote name (pr#17650, Vasu Kulkarni)
  • tests: qa/tests: fix rbd pool creation for systemd tests (pr#17536, Vasu Kulkarni)
  • tests: [qa/tests]: misc ceph-ansible fixes and udpate (pr#17096, Vasu Kulkarni)
  • tests: qa/tests/rados: Remove unsupported 2-size-1-min-size config (pr#17576, Vasu Kulkarni)
  • tests: qa/tests: use ceph-deploy stable branch for single node tests (pr#20979, Vasu Kulkarni)
  • tests: qa/tests: Various whitelists for smoke suite (issue#21376, pr#17680, Vasu Kulkarni)
  • tests: qa/tests: Wip ceph deploy upgrade (pr#17651, Vasu Kulkarni)
  • tests: qa/workunits/rados/test_large_omap_detection: Scrub pgs instead of OSDs (pr#21410, Brad Hubbard)
  • tests: qa/workunits: silence py warnings for ceph-disk tests (issue#22154, pr#19075, Kefu Chai)
  • tests: rados: Copy payload in ceph_perf_msgr_client (issue#22100, pr#18862, Jeegn Chen)
  • tests: rados: Intializing members class StriperTest (pr#16843, amitkuma)
  • tests: remove TestPGLog ASSERT_DEATH test (issue#23504, pr#21117, Nathan Cutler)
  • tests: run-standalone.sh improve error message (pr#17093, David Zafman)
  • tests: run-standalone.sh skip core_pattern if already set (pr#17098, David Zafman)
  • tests: test/admin_socket_output: add –vstart=path/to/asok option (pr#20371, Kefu Chai)
  • tests: test/admin_socket_output: better error reporting (pr#20409, Brad Hubbard)
  • tests: test/admin_socket_output: switch to std::experimental::filesystem (pr#20307, Kefu Chai)
  • tests: test/ceph_test_objectstore: make settings update and restore less error prone (pr#21145, Igor Fedotov)
  • tests: test: checking negative returns from creat() (pr#18090, amitkuma)
  • tests: test/CMakeLists: disable test_pidfile.sh (issue#20975, pr#16977, Sage Weil)
  • tests: test/CMakeLists: disable test-pidfile.sh (pr#17401, Sage Weil)
  • tests: test/coredumpctl: support freebsd (pr#17447, Kefu Chai)
  • tests: test/dashboard: hardcode .coverage path to workaround tox bugs (pr#21485, Kefu Chai)
  • tests: test/dashboard: specify workdir using tox.ini (issue#23709, pr#21416, Kefu Chai)
  • tests: test: Don’t dump core when using EXPECT_DEATH (pr#17390, Kefu Chai)
  • tests: test/fio: extend fio objectstore plugin to better simulate OSD behavior (pr#19101, Igor Fedotov)
  • tests: test/fio: fix building of the fio_ceph_objectstore plugin (pr#18332, Radoslaw Zarzynski)
  • tests: test: Fix and enable test_pidfile.sh (issue#20770, pr#16987, David Zafman)
  • tests: test: Fix ceph-objectstore-tool usage check (pr#17785, David Zafman)
  • tests: test: fix misc fiemap testing (issue#21716, pr#18240, Kefu Chai, Ning Yao)
  • tests: test: Initialization of *comp_racing_read class CopyFromOp (pr#17369, Amit Kumar)
  • tests: test: Initializing ChunkReadOp members (pr#19334, amitkuma)
  • tests: test/journal: Initialize member variable m_work_queue (pr#17089, amitkuma)
  • tests: test/librados: be more tolerant with timed lock tests (issue#20086, pr#20161, Kefu Chai)
  • tests: test/librados: increase pgp_num along with pg_num (issue#23763, pr#21555, Kefu Chai)
  • tests: test/librados: s/invoke_result_t/result_of_t/ (pr#20379, Kefu Chai)
  • tests: test/librados_test_stub: pass snap context to zero op (pr#17186, Mykola Golub)
  • tests: test/log: fix for crash with libc++ (pr#20233, Casey Bodley)
  • tests: test: Make clearer by moving code out of loop (pr#20759, David Zafman)
  • tests: test/objectstore/test_bluefs: cleanups (pr#17909, Kefu Chai)
  • tests: test: only test dashboard_v2 when it is enabled (pr#20777, Willem Jan Withagen)
  • tests: test: only test enabled python bindings (pr#21293, Kefu Chai)
  • tests: test/osd: initialize Non-static class members in WeightedTestGenerator (pr#15922, Jos Collin)
  • tests: test/osd: Non-static class members not initialized in UnsetRedirectOp (pr#15921, Jos Collin)
  • tests: test/osd: silence warnings from -Wsign-compare (pr#17027, Jos Collin)
  • tests: test: put new BlueStore tests un ifdef WITH_BLUESTORE (pr#20576, Willem Jan Withagen)
  • tests: test:qa:infra - Run update daily and use bash (pr#21218, David Galloway)
  • tests: test:qa:infra - teuthology crontab items as of 3/27/18 (pr#21075, Yuri Weinstein)
  • tests: test: reduce the chance to have degraded PGs (issue#22711, pr#20046, Kefu Chai)
  • tests: test: remove distro_version assert in distro detect test (pr#21052, Shengjing Zhu)
  • tests: test: Replace bc command with printf command (pr#21013, David Zafman)
  • tests: test: silence warning from -Wsign-compare (pr#17790, Jos Collin)
  • tests: test: silence warnings from -Wsign-compare (pr#17962, Jos Collin)
  • tests: tests - Replaced requests for “centos 7.3” to centos_latest (pr#19262, Yuri Weinstein)
  • tests: test/store_test: fix FTBFS as Sequencer is removed (pr#20382, Kefu Chai)
  • tests: test/store_test: update Many4KWritesTest* test cases to finalize with… (pr#20230, Igor Fedotov)
  • tests: test/throttle: kill tests exercising dtor of Throttle classes (pr#17442, Kefu Chai)
  • tests: test/unittest_bufferlist: check retvals of syscalls (pr#18238, Kefu Chai)
  • tests: test/unittest_pg_log: silence gcc warning (pr#17328, Kefu Chai)
  • tests: test: Use jq in a compatible way and for easier diff analysis (pr#21450, David Zafman)
  • tests: test: Whitelist corrections (pr#22167, David Zafman)
  • tests,tools: crushtool: print error message to stderr not dout(1) (issue#21758, pr#18242, Kefu Chai)
  • tests: unittest_crypto: Don’t exceed limit for getentropy (pr#18505, Brad Hubbard)
  • tests: vstart: fix initial start when there is no ceph.conf (pr#21019, Jianpeng Ma)
  • The Day Has Come! (pr#19657, Adam C. Emerson)
  • tools: Align use of uint64_t in service_daemon::AttributeType (pr#16938, James Page)
  • tools: ceph-disk: erase 110MB for nuking existing bluestore (issue#22354, pr#20400, Kefu Chai)
  • tools: ceph-disk: fix ‘–runtime’ omission for ceph-osd service (issue#21498, pr#17904, Carl Xiong)
  • tools: ceph-disk: fix signed integer is greater than maximum when call major (pr#19196, Song Shun)
  • tools: ceph-disk: include output of failed command in exception (pr#20497, Kefu Chai)
  • tools: ceph-disk: more precise error message when a disk is specified (pr#18018, Kefu Chai)
  • tools: ceph-disk: reduce the scope of activate_lock (pr#20114, zhaokun)
  • tools: ceph-disk: retry on OSError (issue#21728, pr#18162, Kefu Chai)
  • tools: ceph-disk: unlock all partitions when activate (pr#17363, Kefu Chai)
  • tools: ceph-disk: write log to /var/log/ceph not to /var/run/ceph (pr#18375, Kefu Chai)
  • tools: ceph: fixes for “tell <service>.*” command (issue#21230, pr#17463, Kefu Chai)
  • tools: ceph-kvstore-tool: make it a bit more friendly (pr#21477, Sage Weil)
  • tools: ceph-kvstore-tool: use unique_ptr<> to manage the lifecycle of bluestore (pr#18221, Kefu Chai)
  • tools: ceph-objectstore-tool: Add option “dump-import” to examine an export (issue#22086, pr#19368, David Zafman)
  • tools: ceph-objectstore-tool: Fix set-size to clear data_digest if changing … (pr#18885, David Zafman)
  • tools: ceph-objectstore-tool: “$OBJ get-omaphdr” and “$OBJ list-omap” scan all pgs instead of using specific pg (issue#21327, pr#17985, David Zafman)
  • tools: ceph-objectstore-tool: skip object with generated version (pr#18507, huangjun)
  • tools: ceph-syn: silence clang analyzer warning (pr#18577, Kefu Chai)
  • tools: ceph-volume: Use a delimited CLI output parser instead of JSON (pr#17097, Alfredo Deza)
  • tools: cleanup: rip out ceph-rest-api (issue#21264, issue#22457, pr#17530, Nathan Cutler)
  • tools: correct total size formatting (pr#21641, Zheng Yin)
  • tools: crushtool: add –add-bucket and –move options (pr#20183, Kefu Chai)
  • tools: FreeBSD basic getopt does not use –options (pr#21148, Willem Jan Withagen)
  • tools: Initialization of *server, command variables (pr#17135, amitkuma)
  • tools: make rados get/put/append command help txt clear (issue#22958, pr#20363, lvshuhua)
  • tools: Modify “rados df” header’s alignment (pr#17549, iliul)
  • tools: rados add a cli option to clear omap keys (issue#22255, pr#19180, Abhishek Lekshmanan)
  • tools: rados/tool: fixup rados stat command hint (pr#16983, huanwen ren)
  • tools: script: build-integration-branch: avoid Unicode error (issue#24003, pr#21918, Nathan Cutler)
  • tools: script: ceph-release-notes: minor fixes for split_component (pr#16605, Abhishek Lekshmanan)
  • tools: Special scrub handling of hinfo_key errors (issue#23428, issue#23364, pr#20947, David Zafman)
  • tools: src/vstart.sh: default os to filestore for FreeBSD (pr#17454, xie xingguo)
  • tools: stop.sh: add ceph configure file location (pr#20888, Jianpeng Ma)
  • tools: tools/ceph-conf: dump parsed config in plain text or as json (issue#21862, pr#18350, Piotr Dałek)
  • tools: tools/ceph_monstore_tool: include mgrmap in initial paxos epoch (issue#22266, pr#19780, Kefu Chai)
  • tools: tools/ceph_monstore_tool: rebuild initial mgrmap also (issue#22266, pr#19238, Kefu Chai)
  • tools: tools/ceph-objectstore-tool: command to trim the pg log (issue#23242, pr#20786, Josh Durgin, David Zafman)
  • tools: tools/ceph_objectstore_tool: fix ‘dup’ unable to duplicate meta PG (pr#17572, xie xingguo)
  • tools: tools/rados: improve the ls command usage (pr#21553, Li Wang)
  • tools: tools: rados: make -f be –format for consistency with ceph tool (issue#15904, pr#20147, Nathan Cutler)
  • tools: tools/rados: use the monotonic clock in rados bench (issue#21375, pr#18588, Mohamad Gebai)
  • tools: update monstore tool for fsmap, mgrmap (issue#21577, pr#18005, John Spray)
  • tools: Use –no-mon-config so ceph_objectstore_tool.py test doesn’t hang (pr#21274, David Zafman)
  • tools: vstart.sh: move rgw configuration to client.rgw section (pr#18331, Yan Jun)
  • tools: vstart.sh: use bluestore as default osd objectstore backend (pr#17100, mychoxin)
  • vstart: fix option (due to quotes) and allow disabling dashboard (issue#23345, pr#20986, Joao Eduardo Luis)
  • vstart.sh: fix a typo (pr#18729, iliul)
  • vstart.sh: Fix help text in vstart.sh (pr#21071, Marc Koderer)
  • vstart.sh: quote cmd params when display executing cmd (pr#17057, Jiaying Ren)
  • vstart.sh: quote command only when necessary (pr#18181, Kefu Chai)
  • vstart.sh: should quote the parameters to get them quoted (pr#18523, Kefu Chai)
  • vstart.sh: simplify the objectstore related logic (pr#17749, Kefu Chai)