openSUSE Build Service

Overview Details Repositories Revisions Requests Users Attributes Meta

File _patchinfo of Package patchinfo (Revision d5b70839882b509bdb71f7f75eee9138)

Currently displaying revision d5b70839882b509bdb71f7f75eee9138 , Show latest

<patchinfo incident="6051">
  <issue id="987144" tracker="bnc">VUL-0: CVE-2016-5009: ceph:  moncommand with empty prefix crashes monitor</issue>
  <issue id="999688" tracker="bnc">ceph-deploy mon create-initial fails on one node</issue>
  <issue id="985232" tracker="bnc">MDS dies while running xfstests generic/013 - FAILED assert(in-&gt;is_file())</issue>
  <issue id="1008501" tracker="bnc">TypeError: _recover_auth_meta() takes exactly 2 arguments</issue>
  <issue id="1014338" tracker="bnc">L3: OSD daemon uses ~100% CPU load right after OSD creation / first OSD start</issue>
  <issue id="990438" tracker="bnc">civetweb HTTPS support not working</issue>
  <issue id="1007216" tracker="bnc">systemd is killing off OSDs ("cgroup: fork rejected by pids controller")</issue>
  <issue id="977940" tracker="bnc">"make check" fails due to linker errors when building unit tests with --with-xio</issue>
  <issue id="982141" tracker="bnc">ceph is in unresolvable state for ppc64le</issue>
  <issue id="1005179" tracker="bnc">Performance fix is missing in SES4</issue>
  <issue id="1008894" tracker="bnc">Error EINVAL: invalid command in SOC7 with ceph</issue>
  <issue id="2016-5009" tracker="cve" />
  <category>security</category>
  <rating>moderate</rating>
  <packager>smithfarm</packager>
  <description>
This update for ceph fixes the following issues:

- Update to version 10.2.4+git.1481215985.12b091b:
  + msg/simple/Pipe: avoid returning 0 on poll timeout (boo#1014338)

- Update to version 10.2.3+git.1481191611.132dac2:
Upstream 10.2.4 point release:
  + os/filestore/HashIndex: fix list_by_hash_* termination on reaching end
  + osd/osd_types: conditional pg_pool_t encoding
  + osd/osd_types: encode pg_pool_t like hammer if features indicate hammer
  + crush/CrushWrapper: drop unused 'lean' encode() argument
  + crush/CrushWrapper: encode with features
  + crush: condition latest tunable encoding on features
  + librados: remove new setxattr overload to avoid breaking the C++ ABI
  + mon: MonmapMonitor: return success when monitor will be removed
  + mon: MonmapMonitor: drop unnecessary 'goto' statements
  + OSDMonitor: only reject MOSDBoot based on up_from if inst matches
  + tests: adapt rbd CLI help test for SES3/4
Downstream-only commits:
  + build/ops: fix undefined crypto references with --with-xio (boo#977940)
  + ceph.spec.in: make check only in ceph-test package (for automating "make
    check" in the build service)

- Update to version 10.2.3+git.1479299521.4429782:
Pull commits from upstream "jewel" branch (which will very likely be released
soon as Ceph 10.2.4):
  + ceph_test_msgr: use ceph log infrastructure to output
  + test/ceph_test_msgr: fix circular locking dependency
  + test/ceph_test_msgr: do not use Message::middle for holding transient data
  + all: add const to operator&lt;&lt; param
  + messages/MForward: fix encoding features
  + messages/MForward: reencode forwarded message if target has differing features
  + msg/Message: fix set_middle vs throttler
  + msg: adjust byte_throttler from Message::encode
  + mon/OSDMonitor: encode canonical full osdmap based on osdmap flags
  + mon/OSDMonitor: health warn if require_{jewel,kraken} flags aren't set
  + mon/OSDMonitor: encode OSDMap::Incremental with same features as OSDMap
  + mon: expose require_jewel_osds flag to user
  + ceph-create-keys: add missing argument comma
Reinstate upstream NFS backports from https://github.com/ceph/ceph/pull/11335
to support inclusion of NFS-Ganesha V2.4.1 in SES4 product release:
  + Revert "Revert "rgw_file: fix rename cases and unify unlink""
  + Revert "Revert "rgw_file: refuse partial, out-of-order writes""
  + Revert "Revert "rgw_file: implement rgw_setattr""
  + Revert "Revert "rgw_file: refuse ops on deleted targets""
  + Revert "Revert "rgw file: refuse to rename directories""
  + Revert "Revert "rgw nfs: don't leak fh-&gt;mtx in lookup_handle()""
  + Revert "Revert "rgw nfs: fix write continuations""
  + Revert "Revert "rgw_file: unlock() must precede out label""
  + Revert "Revert "rgw file: remove busy-wait in RGWLibFS::gc()""
  + Revert "Revert "rgw_file: fix/update rgw_open flag arguments""
  + Revert "Revert "rgw_file: fix/update rgw_create flag arguments""
  + Revert "Revert "rgw_file: implement stateless write completion timer""
  + Revert "Revert "rgw_file: add RGW_OPEN_FLAG_STATELESS as a synonym for V3""
  + Revert "Revert "rgw_file:  log on write completion events at level 10""
  + Revert "Revert "rgw file: allow RGWLibFS::write_completion_interval_s to be set in conf""
  + Revert "Revert "librgw: add API version defines for librgw and rgw_file""
  + Revert "Revert "ceph_timer: prefer using directive to typedef""
  + Revert "Revert "rgw_file:  explain semantics of RGWFileHandle::commit""
  + Revert "Revert "rgw_file: restore local definition of RGWLibFS gc interval""
  + Revert "Revert "rgw: fix the wrong return value""
  + Revert "Revert "rgw: rgw file fix bug of rgw_lookup can not exact match file name""
  + Revert "Revert "rgw: remove duplicated calls to getattr""
  + Revert "Revert "rgw_file: implement rgw_setattr""
  + Revert "Revert "rgw_file: fix set_attrs operation""
  + Revert "Revert "rgw_file:  pre-assign times""

- Update to version 10.2.3+git.1478701510.3f36d65:
Revert upstream NFS backports from https://github.com/ceph/ceph/pull/11335
because they break nfs-ganesha 2.3
  + Revert "rgw_file:  pre-assign times"
  + Revert "rgw_file: fix set_attrs operation"
  + Revert "rgw_file: implement rgw_setattr"
  + Revert "rgw: remove duplicated calls to getattr"
  + Revert "rgw: rgw file fix bug of rgw_lookup can not exact match file name"
  + Revert "rgw: fix the wrong return value"
  + Revert "rgw_file: restore local definition of RGWLibFS gc interval"
  + Revert "rgw_file:  explain semantics of RGWFileHandle::commit"
  + Revert "ceph_timer: prefer using directive to typedef"
  + Revert "librgw: add API version defines for librgw and rgw_file"
  + Revert "rgw file: allow RGWLibFS::write_completion_interval_s to be set in conf"
  + Revert "rgw_file:  log on write completion events at level 10"
  + Revert "rgw_file: add RGW_OPEN_FLAG_STATELESS as a synonym for V3"
  + Revert "rgw_file: implement stateless write completion timer"
  + Revert "rgw_file: fix/update rgw_create flag arguments"
  + Revert "rgw_file: fix/update rgw_open flag arguments"
  + Revert "rgw file: remove busy-wait in RGWLibFS::gc()"
  + Revert "rgw_file: unlock() must precede out label"
  + Revert "rgw nfs: fix write continuations"
  + Revert "rgw nfs: don't leak fh-&gt;mtx in lookup_handle()"
  + Revert "rgw file: refuse to rename directories"
  + Revert "rgw_file: refuse ops on deleted targets"
  + Revert "rgw_file: implement rgw_setattr"
  + Revert "rgw_file: refuse partial, out-of-order writes"
  + Revert "rgw_file: fix rename cases and unify unlink"

- Update to version 10.2.3+git.1478637374.b4c314a:
  + rgw: fix for bucket delete racing with mdlog sync

- Update to version 10.2.3+git.1478595853.bfb2d58:
  + ceph-create-keys: add missing argument comma (boo#1008894)

- Update to version 10.2.3+git.1478513749.4d4db79:
  + common: Improve linux dcache hash algorithm (boo#1005179)

- Update to version 10.2.3+git.1478275751.8c1c654:
New commits from upstream jewel (pre-10.2.4) branch
  + mon : Display full flag in ceph status if full flag is set
  + osd: add peer_addr in heartbeat_check log message
  + msg/msg_types: update sockaddr, sockaddr_storage accessors
  + client: kill QuotaTree
  + common: fix value of CINIT_FLAG_DEFER_DROP_PRIVILEGES
  + crush: reset bucket-&gt;h.items[i] when removing tree item
  + librbd: journal::Replay no longer holds lock while completing callback
  + librbd: prevent creation of clone from non-primary mirrored image
  + librbd: prevent creation of v2 image ids that are too large
  + rgw: fix potential memory leaks in RGWPutCORS_ObjStore_S3::get_params
  + mon/osdmonitor: decouple adjust_heartbeat_grace and min_down_reporters
  + ceph-disk: timeout ceph-disk to avoid blocking forever
  + mon/osdmonitor: initialize local variable "kb_avail_i"
  + mon/osdmonitor: fix incorrect output of "osd df" due to osd out
  + fast dispatch heartbeat message
  + librbd: request exclusive lock if current owner cannot execute op
  + LTTng-UST disabled for openSUSE
  + rbd: bench io-size should not be larger than image size
  + ceph-osd-prestart.sh: check existence of OSD data directory
  + Revert "common: add int64_t template for strict_si_cast()"
  + common/config: cast OPT_U32 options using uint32_t
  + common: add int64_t template for strict_si_cast()
  + common: instantiate strict_si_cast&lt;long&gt; not strict_si_cast&lt;int64_t&gt;
  + librbd: fix possible inconsistent state when disabling mirroring
  + mon/PGMonitor: mark dump_object_stat_sum() as static
  + mon/PGMonitor: calc the %USED of pool using used/(used+avail)
  + crush: When consider rule size use the correct method.
  + rpm: proper packaging of udev directories
  + ObjectCacher: fix last_write check in bh_write_adjacencies()
  + rgw: fix period update --commit return error
  + rgw: fix radosgw daemon core when reopen logs
  + client: add mask parameter to _do_lookup
  + client: plumb a mask argument into _lookup
  + client: only skip querying the MDS in _lookup when we have the necessary caps
  + ceph: don't fudge the ctime in stat() unless it's really older than the mtime
  + PG: introduce and maintain upset
  + PG: use upset rather than up for _update_calc_stats
  + ceph-disk: do not activate device that is not ready
  + Cleanup: delete find_best_info again
  + ReplicatedPG: call op_applied for submit_log_entries based repops
  + rgw file: remove busy-wait in RGWLibFS::gc()
  + rgw_file: restore local definition of RGWLibFS gc interval
  + rgw: fix upgrade from old multisite to new multisite configuration
  + client: add missing client_lock for get_root
  + client: fix shutdown with open inodes
  + qa: rbd/map-snapshot-io.sh: don't chown sysfs files
  + qa: rbd: don't modprobe, chown sysfs files or udevadm settle
  + qa: rbd/concurrent.sh: suppress rbd map output
  + qa: rbd/simple_big.sh: drop extraneous sudo
  + krbd: don't segfault if images are unmapped concurrently
  + doc: add "--orphan-stale-secs" to radosgw-admin(8)
  + radosgw-admin: add "--orphan-stale-secs" to --help
  + Add two options to radosgw-admin.rst manpage
  + Add Install section to systemd rbdmap.service file
  + doc: fix description for rsize and rasize
  + librados: modify Pipe::connect() to return the error code
  + log: Log.cc: Assign LOG_INFO priority to syslog calls
  + logrotate: Run as root/ceph
  + mon: OSDMonitor: Missing nearfull flag set
  + hobject: clarify is_snap and has_snapset for max
  + src/osd: relax the requirement that we scrub a whole hash value
  + osd_types,PG: force the map used in _scrub to be sorted correctly
  + PG: block writes and scan log for scrub using inclusive upper bound
  + rgw: RGWDataSyncCR fails on errors from RGWListBucketIndexesCR
  + rgw: add reinit/rebind logic (ldap)
  + rgw ldap: enforce simple_bind w/LDAPv3
  + rgw ldap:  protect rgw::from_base64 from non-base64 input
  + add ldap auth custom search filter feature patch - http://tracker.ceph.com/issues/17185
  + rgw_file: fix rename cases and unify unlink
  + rgw_file: refuse partial, out-of-order writes
  + rgw_file: implement rgw_setattr
  + rgw_file: refuse ops on deleted targets
  + rgw file: refuse to rename directories
  + rgw nfs: don't leak fh-&gt;mtx in lookup_handle()
  + rgw nfs: fix write continuations
  + rgw_file: unlock() must precede out label
  + rgw file: remove busy-wait in RGWLibFS::gc()
  + rgw_file: fix/update rgw_open flag arguments
  + rgw_file: fix/update rgw_create flag arguments
  + rgw_file: implement stateless write completion timer
  + rgw_file: add RGW_OPEN_FLAG_STATELESS as a synonym for V3
  + rgw_file:  log on write completion events at level 10
  + rgw file: allow RGWLibFS::write_completion_interval_s to be set in conf
  + librgw: add API version defines for librgw and rgw_file
  + ceph_timer: prefer using directive to typedef
  + rgw_file:  explain semantics of RGWFileHandle::commit
  + rgw_file: restore local definition of RGWLibFS gc interval
  + rgw: fix the wrong return value
  + rgw: rgw file fix bug of rgw_lookup can not exact match file name
  + rgw: remove duplicated calls to getattr
  + rgw_file: implement rgw_setattr
  + rgw_file: fix set_attrs operation
  + rgw_file:  pre-assign times
  + doc/radosgw: fix description of response elements 'Part'
  + librbd: new journal policy to disable initializing the journal
  + librbd: permit disabling journaling if in corrupt state
  + rgw: set args when rebuilding req_info
  + rgw: forward input data when forwarding set_bucket_version to master
  + init-radosgw: do not use systemd-run in sysvinit
  + messages: fix MMDSBeacon constructor
  + mon: don't crash on invalid standby_for_fscid
  + mds: make mds_role_t member order match constructor
  + mds: remove max_mds config option
  + os/filestore/FileJournal: fail out if FileJournal is not block device or regular file
  + cephx: Fix multiple segfaults due to attempts to encrypt or decrypt an
    empty secret and a null CryptoKeyHandler
  + common: only call crypto::init once per CephContext
  + mds: catch duplicates in DamageTable
  + client: fix segment fault in Client::_invalidate_kernel_dcache().
  + mds: remove fail-safe queueing replay request
  + client: fix readdir vs fragmentation race
  + ceph-create-keys: fix existing-but-different case
  + doc: clarify rbd size units
  + rbd: recognize lock_on_read option
  + rbd: fix up terminology in help texts
  + rbd: expose rbd unmap options
  + mds: s/used_ions/used_inos/
  + mds: log path with CDir damage messages
  + client: properly set inode number of created inode in replay request
  + librbd: ignore notify errors on missing image header
  + build: include more files in "make dist" tarball
  + cls/journal: add async client_update_state method
  + journal: allow to trim journal for "laggy" clients
  + librbd: optionally flag "laggy" journal clients disconnected
  + rbd: new command to disconnect journal client
  + rbd-mirror: decode_client_meta should return false on error
  + rbd-mirror: resync was possible only when image replayer start had succeeded
  + rbd-mirror: stop replay when client is disconnected
  + rbd-mirror: option to automatically resync after journal client disconnect
  + qa/workunits/rbd: demote/promote image on same cluster
  + rbd-mirror: include tag tid in bootstrap debug log messages
  + librbd: separate journal::TagData predecessor data into new struct
  + librbd: block RPC requests when demoting the image
  + librbd: normalize journal promotion/demotion events
  + rbd-mirror: demote/promote in same cluster results in split-brain
  + rbd-mirror: prevent syncing to non-primary image after failover
  + test: fixed unused function warnings in unittest_rbd_mirror
  + librbd: fix journal create compilation error
  + qa/workunits/rbd: before removing image make sure it is not bootstrapped
  + test/rbd: fix possible mock journal race conditions
  + qa/workunits/rbd: new mirroring forced promotion test
  + qa/workunits/rbd: fix remove mirrored image race conditions
  + journal: expose ability to retrieve partial tag list
  + librbd: unify journal event callbacks into single interface
  + librbd: helper class for quiescing in-flight async ops
  + librbd: new journal listener event for force promotion
  + journal: send update notification when tag allocated
  + journal: increase concurrency of journal recorder
  + journal: make librados call async in ObjectRecorder
  + rbd: Fix race between journal flush and append events
  + rbd-mirror: test: Fixed timeout problem in rbd_mirror_stress.sh
  + rbd-mirror: concurrent access of event might result in heap corruption
  + rbd-mirror: use pool id + global image id as deletion primary key
  + librbd: support deleting image by id instead of name
  + common: move AsyncOpTracker to common library
  + journal: fix AsyncOpTracker compiler error
  + rbd-mirror: wait for in-flight event commit notifications
  + journal: avoid holding lock while sending journal append
  + journal: use reverse iterator search to find flush record
  + journal: delay moving overflowed buffers until quiesced
  + journal: clean up object recorder closed/overflow callback
  + librbd: ignore attempts to request lock if already lock owner
  + rbd-mirror: potential race allocating tag during shut down
  + qa/workunits: reduce amount of client IO for rbd-mirror stress
  + journal: complete action only after notification completed
  + journal: ensure in-flight ops are complete destroying journaler
  + rgw: Let the default quota settings take effect during user creation
  + rgw: Do not bother to check against conf values during quota handling
  + rgw: Add documentation for RGW default quota
  + rbd: restrict mirror enable/disable actions on parents/clones
  + librbd: enable/disable of features is not allowed when already enabled/disabled
  + rbd: cleanup - Proxied operations shouldn't result in error messages if replayed
  + cls/rbd: set_features should adjust the mask to known features
  + librbd: update_features should handle Infernalis OSDs
  + librbd: potential deadlock closing image with in-flight readahead
  + librbd: block name prefix might overflow fixed size C-string
  + librbd: possible deadlock if cluster connection closed after image
  + librbd: corrected use-after-free in ImageWatcher
  + rgw:user email can modify to empty
  + rgw: remove user email index using helper function
  + rgw: don't fail if lost race when setting acls
  + rgw: fix versioned object IO error
  + doc: Radosgw multitenancy config paramater change
  + rgw:response information is error when geting token of swift account The
    header is X-Trans-Id instead of x-amz-request-id in the response header
  + doc: Modification for "TEST S3 ACCESS" section
  + Allow building RGW with LDAP disabled
  + tests: populate /dev/disk/by-partuuid for scsi_debug
  + AArch64: Detect crc32 extension support from assembler
  + rgw: mark configuration as converted to avoid corruption when running older admins
  + rgw: Fix missing master zone for default region conversion
  + rgw: avoid corruption when running old radosgw-admin on a newer rgw
  + mon/AuthMonitor: make AuthMonitor::IncType public
  + tools/ceph-objectstore-tool: add "update-mon-db" command
  + tools/ceph_monstore_tool: add "rebuild" command
  + doc: add rados/operations/disaster-recovery.rst
  + tools/rebuild_mondb: avoid unnecessary result code cast
  + tools/rebuild_mondb: kill compiling warning
  + tools/ceph_monstore_tool: update pgmap_meta also when rebuilding store.db
  + tools/ceph_monstore_tool: bail out if no caps found for a key
  + doc: fill keyring with caps before passing it to ceph-monstore-tool
  + tests: ceph-disk: force debug monc = 0
  + rgw: fix regression with handling double underscore
  + rgw: set correct instance on the object
  + rbd-nbd: fix kernel deadlock during teuthology testing
  + rbd-nbd: mask out-of-bounds IO errors caused by image shrink
  + librbd: invalidate cache before trimming image
  + librbd: ignore cache busy errors when shrinking an image
  + test: TestJournalReplay test cases need to wait for committed journal event
  + Don't loop forever when reading data from 0 sized segment.
  + journal: do not prematurely flag object recorder as closed
  + mon: fix missing osd metadata (again)
  + rgw: Do not archive metadata by default
  + ceph-disk: change ownership of init file to ceph:ceph
  + ceph-disk: Compatibility fixes for Python 3
  + ceph-disk: Use true integer division in get_dev_size
  + ceph-disk: Fix bug in FileLock
  + ceph-disk: Use context manager with FileLock
  + ceph-disk: Use os.path.isabs instead of .startswith('/')
  + ceph-disk: Set space_symlink to the path, not file object
  + ceph-disk: PEP8ify
  + ceph-disk: allow using a regular file as a journal
  + rgw: handle empty POST condition
  + librbd: discard after write can result in assertion failure
  + qa: remove EnumerateObjects from librados upgrade tests
  + rgw: fix put_acls for objects starting and ending with underscore
  + ceph-post-file: migrate to RSA SSH keys
  + ceph-post-file: Ignore keys offered by ssh-agent
Commits necessary to resolve merge conflicts
  + Revert "ceph_disk.main: s/get_dev_size/get_free_partition_size/"
  + build/ops: include man pages in tarball explicitly
Fix boo#1007216
  + Revert "Revert "Drop any systemd imposed process/thread limits""
Fix boo#1008501
  + ceph_volume_client: fix _recover_auth_meta() method
  + ceph_volume_client: check if volume metadata is empty
  + ceph_volume_client: fix partial auth recovery

- fix "ceph --version" so it reports 10.2.3
- add changes file entry for bugfix included in upstream 10.2.3 release:
  VUL-0: CVE-2016-5009: ceph: moncommand with empty prefix crashes monitor
  (#16297, boo#987144)

- ibs_constraints.sh: bash script to update _constraints for IBS
- pre_checkin.sh: add ibs_constraints.sh to ceph-test Sources list

- Update to version 10.2.3+git.1475228057.755cf99:
Upstream 10.2.3 release
* build/ops: 60-ceph-partuuid-workaround-rules still needed by debian jessie
  (udev 215-17) (#16351)
* build/ops: ceph Resource Agent does not work with systemd (#14828)
* build/ops: ceph-base requires parted (#16095)
* build/ops: ceph-osd-prestart.sh contains Upstart-specific code (#15984)
* build/ops: mount.ceph: move from ceph-base to ceph-common and add symlink in
  /sbin for SUSE (#16598, #16645)
* build/ops: need rocksdb commit 7ca731b12ce for ppc64le build (#17092, boo#982141)
* build/ops: rpm: OBS needs ExclusiveArch (#16936)
* cli: ceph command line tool chokes on ceph &#8211;w (the dash is unicode 'en dash'
  &amp;ndash, copy-paste to reproduce) (#12287)
* common: expose buffer const_iterator symbols (#16899, Noah Watkins)
* common: global-init: fixup chown of the run directory along with log and asok 
  files (#15607)
* fs: ceph-fuse: link to libtcmalloc or jemalloc (#16655)
* fs: client: crash in unmount when fuse_use_invalidate_cb is enabled (#16137)
* fs: client: fstat cap release (#15723)
* fs: essential backports for OpenStack Manila (#15406, #15614, #15615)
* fs: fix double-unlock on shutdown (#17126)
* fs: fix mdsmap print_summary with standby replays (#15705)
* fs: fuse mounted file systems fails SAMBA CTDB ping_pong rw test with v9.0.2
  (#12653, #15634)
* librados: Add cleanup message with time to rados bench output (#15704)
* librados: Missing export for rados_aio_get_version in src/include/rados/librados.h
  (#15535)
* librados: osd: bad flags can crash the osd (#16012)
* librbd: Close journal and object map before flagging exclusive lock as released 
  (#16450)
* librbd: Crash when utilizing advisory locking API functions (#16364)
* librbd: ExclusiveLock object leaked when switching to snapshot (#16446)
* librbd: FAILED assert(object_no &lt; m_object_map.size()) (#16561)
* librbd: Image removal doesn't necessarily clean up all rbd_mirroring entries 
  (#16471)
* librbd: Object map/fast-diff invalidated if journal replays the same snap
  remove event (#16350)
* librbd: Timeout sending mirroring notification shouldn't result in failure
  (#16470)
* librbd: Whitelist EBUSY error from snap unprotect for journal replay (#16445)
* librbd: cancel all tasks should wait until finisher is done (#16517)
* librbd: delay acquiring lock if image watch has failed (#16923)
* librbd: fix missing return statement if failed to get mirror image state (#16600)
* librbd: flag image as updated after proxying maintenance op (#16404)
* librbd: mkfs.xfs slow performance with discards and object map (#16707, #16689)
* librbd: potential use after free on refresh error (#16519)
* librbd: rbd-nbd does not properly handle resize notifications (#15715)
* librbd: the option 'rbd_cache_writethrough_until_flush=true' doesn't work 
  (#16740, #16386, #16708, #16654, #16478)
* mds:  tell command blocks forever with async messenger (TestVolumeClient.
  test_evict_client failure) (#16288)
* mds: Confusing MDS log message when shut down with stalled journaler reads 
  (#15689)
* mds: Deadlock on shutdown active rank while busy with metadata IO (#16042)
* mds: Failing file operations on kernel based cephfs mount point leaves
  unaccessible file behind on hammer 0.94.7 (#16625, boo#985232)
* mds: Fix shutting down mds timed-out due to deadlock (#16396)
* mds: MDSMonitor fixes (#16136)
* mds: MDSMonitor::check_subs() is very buggy (#16022)
* mds: Session::check_access() is buggy (#16358)
* mds: StrayManager.cc: 520: FAILED assert(dnl-&gt;is_primary()) (#15920)
* mds: enforce a dirfrag limit on entries (#16164)
* mds: fix SnapRealm::have_past_parents_open() (#16299)
* mds: fix getattr starve setattr (#16154)
* mds: wrongly treat symlink inode as normal file/dir when symlink inode is stale 
  on kcephfs (#15702)
* mon: "mon metadata" fails when only one monitor exists (#15866)
* mon: Monitor: validate prefix on handle_command() (#16297)
* mon: OSDMonitor: drop pg temps from not the current primary (#16127)
* mon: prepare_pgtemp needs to only update up_thru if newer than the existing one 
  (#16185)
* msgr: AsyncConnection::lockmsg/async lockdep cycle: AsyncMessenger::lock,
  MDSDaemon::mds_lock, AsyncConnection::lock (#16237)
* msgr: async messenger mon crash (#16378, #16418)
* msgr: backports of all asyncmsgr fixes to jewel (#15503, #15372)
* msgr: msg/async: connection race hang (#15849)
* osd: FileStore: umount hang because sync thread doesn't exit (#15695)
* osd: Fixes for list-inconsistent-* (#15766, #16192, #15719)
* osd: New pools have bogus stuck inactive/unclean HEALTH_ERR messages until
  they are first active and clean (#14952)
* osd: OSD crash with Hammer to Jewel Upgrade: void FileStore::init_temp_collections()
  (#16672)
* osd: OSD failed to subscribe skipped osdmaps after ceph osd pause (#17023)
* osd: ObjectCacher split BufferHead read fix (#16002)
* osd: ReplicatedBackend doesn't increment stats on pull, only push (#16277)
* osd: Scrub error: 0/1 pinned (#15952)
* osd: crash adding snap to purged_snaps in ReplicatedPG::WaitingOnReplicas (#15943)
* osd: partprobe intermittent issues during ceph-disk prepare (#15176)
* osd: saw valgrind issues in ReplicatedPG::new_repop (#16801)
* osd: sparse_read on ec pool should return extends with correct offset (#16138)
* osd:sched_time not actually randomized (#15890)
* rbd: ImageReplayer::is_replaying does not include flush state (#16970)
* rbd: Journal duplicate op detection can cause lockdep error (#16363)
* rbd: Journal needs to handle duplicate maintenance op tids (#16362)
* rbd: Unable to disable journaling feature if in unexpected mirror state (#16348)
* rbd: bashism in src/rbdmap (#16608)
* rbd: doc: format 2 now is the default image format (#17026)
* rbd: hen journaling is enabled, a flush request shouldn't flush the cache (#15761)
* rbd: possible race condition during journal transition from replay to ready
  (#16198)
* rbd: qa/workunits/rbd: respect RBD_CREATE_ARGS environment variable (#16289)
* rbd: rbd-mirror should disable proxied maintenance ops for non-primary image 
  (#16411)
* rbd: rbd-mirror: FAILED assert(m_local_image_ctx-&gt;object_map != nullptr) (#16558)
* rbd: rbd-mirror: FAILED assert(m_on_update_status_finish == nullptr) (#16956)
* rbd: rbd-mirror: FAILED assert(m_state == STATE_STOPPING) (#16980)
* rbd: rbd-mirror: ensure replay status formatter has completed before stopping
  replay (#16352)
* rbd: rbd-mirror: include local pool id in resync throttle unique key (#16536,
  #15239, #16488, #16491, #16329, #15108, #15670)
* rbd: rbd-mirror: potential race condition accessing local image journal (#16230)
* rbd: rbd-mirror: reduce memory footprint during journal replay (#16321,
  #16489, #16622, #16539, #16223, #16349)
* rgw: A query on a static large object fails with 404 error (#16015)
* rgw: Add zone rename to radosgw_admin (#16934)
* rgw: Bucket index shards orphaned after bucket delete (#16412)
* rgw: Bug when using port 443s in rgw (#16548)
* rgw: Fallback to Host header for bucket name (#15975)
* rgw: Fix civetweb IPv6 (#16928)
* rgw: Increase log level for messages occuring while running rgw admin command
  (#16935)
* rgw: No Last-Modified, Content-Size and X-Object-Manifest headers if no
  segments in DLO manifest (#15812)
* rgw: RGWPeriodPuller tries to pull from itself (#16939)
* rgw: Set Access-Control-Allow-Origin to a Asterisk if allowed in a rule (#15348)
* rgw: Swift API returns double space usage and objects of account metadata (#16188)
* rgw: account/container metadata not actually present in a request are deleted
  during POST through Swift API (#15977, #15779)
* rgw: add socket backlog setting for via ceph.conf (#16406)
* rgw: add tenant support to multisite sync (#16469, #16121, #16665)
* rgw: add_zone only clears master_zone if --master=false (#15901)
* rgw: aws4 parsing issue (#15940, #15939)
* rgw: aws4: add STREAMING-AWS4-HMAC-SHA256-PAYLOAD support (#16146)
* rgw: backport merge of static sites fixes (#15555, #15532, #15531)
* rgw: can set negative max_buckets on RGWUserInfo (#14534)
* rgw: cleanup radosgw-admin temp command as it was deprecated (#16023)
* rgw: comparing return code to ERR_NOT_MODIFIED in rgw_rest_s3.cc (needs minus
  sign) (#16327)
* rgw: custom metadata aren't camelcased in Swift's responses (#15902)
* rgw: data sync stops after getting error in all data log sync shards (#16530)
* rgw: default zone and zonegroup cannot be added to a realm (#16839)
* rgw: document multi tenancy (#16635)
* rgw: don't unregister request if request is not connected to manager (#15911)
* rgw: failed to create bucket after upgrade from hammer to jewel (#16627)
* rgw: fix ldap bindpw parsing (#16286)
* rgw: fix multi-delete query param parsing (#16618)
* rgw: improve support for Swift's object versioning (#15925)
* rgw: initial slashes are not properly handled in Swift's BulkDelete (#15948)
* rgw: master: build failures with boost &gt; 1.58 (#16392, #16391)
* rgw: multisite segfault on ~RGWRealmWatcher if realm was deleted (#16817)
* rgw: multisite sync races with deletes (#16222, #16464, #16220, #16143)
* rgw: multisite: preserve zone's extra pool (#16712)
* rgw: object expirer's hints might be trimmed without processing in some 
  circumstances (#16705, #16684)
* rgw: radosgw-admin failure for user create after upgrade from hammer to jewel 
  (#15937)
* rgw: radosgw-admin: EEXIST messages for create operations (#15720)
* rgw: radosgw-admin: inconsistency in uid/email handling (#13598)
* rgw: realm pull fails when using apache frontend (#15846)
* rgw: retry on bucket sync errors (#16108)
* rgw: s3website: x-amz-website-redirect-location header returns malformed HTTP
  response (#15531)
* rgw: segfault in RGWOp_MDLog_Notify (#16666)
* rgw: segmentation fault on error_repo in data sync (#16603)
* rgw: selinux denials in RGW (#16126)
* rgw: support size suffixes for --max-size in radosgw-admin command (#16004)
* rgw: updating CORS/ACLs might not work in some circumstances (#15976)
* rgw: use zone endpoints instead of zonegroup endpoints (#16834)
* tests: improve rbd-mirror test case coverage (#16197)
* tests: rados/test.sh workunit timesout on OpenStack (#15403)
* tools: ceph-disk: Accept bcache devices as data disks (#13278)
* tools: src/script/subman fails with KeyError: 'nband' (#16961)

- Update to version 10.2.2+git.1475131634.4dc33de:
  + Revert "rpm: drop obsolete python-ceph-compat metapackage"
  + AArch64: Detect crc32 extension support from assembler (boo#999688)

- Update to version 10.2.2+git.1472752588.dada3f3:
  + spec: add ssl requirements
  + rgw: blanket apply NO_SSL_DL to civetweb sections
  (boo#990438)

- Update to version 10.2.4+git.1481215985.12b091b:
  + msg/simple/Pipe: avoid returning 0 on poll timeout (boo#1014338)

- Update to version 10.2.3+git.1478637374.b4c314a:
  + rgw: fix for bucket delete racing with mdlog sync

- Update to version 10.2.3+git.1478595853.bfb2d58:
  + ceph-create-keys: add missing argument comma (boo#1008894)

- Update to version 10.2.3+git.1478513749.4d4db79:
  + common: Improve linux dcache hash algorithm (boo#1005179)

- Update to version 10.2.3+git.1478275751.8c1c654:
New commits from upstream jewel (pre-10.2.4) branch
  + mon : Display full flag in ceph status if full flag is set
  + osd: add peer_addr in heartbeat_check log message
  + msg/msg_types: update sockaddr, sockaddr_storage accessors
  + client: kill QuotaTree
  + common: fix value of CINIT_FLAG_DEFER_DROP_PRIVILEGES
  + crush: reset bucket-&gt;h.items[i] when removing tree item
  + librbd: journal::Replay no longer holds lock while completing callback
  + librbd: prevent creation of clone from non-primary mirrored image
  + librbd: prevent creation of v2 image ids that are too large
  + rgw: fix potential memory leaks in RGWPutCORS_ObjStore_S3::get_params
  + mon/osdmonitor: decouple adjust_heartbeat_grace and min_down_reporters
  + ceph-disk: timeout ceph-disk to avoid blocking forever
  + mon/osdmonitor: initialize local variable "kb_avail_i"
  + mon/osdmonitor: fix incorrect output of "osd df" due to osd out
  + fast dispatch heartbeat message
  + librbd: request exclusive lock if current owner cannot execute op
  + LTTng-UST disabled for openSUSE
  + rbd: bench io-size should not be larger than image size
  + ceph-osd-prestart.sh: check existence of OSD data directory
  + Revert "common: add int64_t template for strict_si_cast()"
  + common/config: cast OPT_U32 options using uint32_t
  + common: add int64_t template for strict_si_cast()
  + common: instantiate strict_si_cast&lt;long&gt; not strict_si_cast&lt;int64_t&gt;
  + librbd: fix possible inconsistent state when disabling mirroring
  + mon/PGMonitor: mark dump_object_stat_sum() as static
  + mon/PGMonitor: calc the %USED of pool using used/(used+avail)
  + crush: When consider rule size use the correct method.
  + rpm: proper packaging of udev directories
  + ObjectCacher: fix last_write check in bh_write_adjacencies()
  + rgw: fix period update --commit return error
  + rgw: fix radosgw daemon core when reopen logs
  + client: add mask parameter to _do_lookup
  + client: plumb a mask argument into _lookup
  + client: only skip querying the MDS in _lookup when we have the necessary caps
  + ceph: don't fudge the ctime in stat() unless it's really older than the mtime
  + PG: introduce and maintain upset
  + PG: use upset rather than up for _update_calc_stats
  + ceph-disk: do not activate device that is not ready
  + Cleanup: delete find_best_info again
  + ReplicatedPG: call op_applied for submit_log_entries based repops
  + rgw file: remove busy-wait in RGWLibFS::gc()
  + rgw_file: restore local definition of RGWLibFS gc interval
  + rgw: fix upgrade from old multisite to new multisite configuration
  + client: add missing client_lock for get_root
  + client: fix shutdown with open inodes
  + qa: rbd/map-snapshot-io.sh: don't chown sysfs files
  + qa: rbd: don't modprobe, chown sysfs files or udevadm settle
  + qa: rbd/concurrent.sh: suppress rbd map output
  + qa: rbd/simple_big.sh: drop extraneous sudo
  + krbd: don't segfault if images are unmapped concurrently
  + doc: add "--orphan-stale-secs" to radosgw-admin(8)
  + radosgw-admin: add "--orphan-stale-secs" to --help
  + Add two options to radosgw-admin.rst manpage
  + Add Install section to systemd rbdmap.service file
  + doc: fix description for rsize and rasize
  + librados: modify Pipe::connect() to return the error code
  + log: Log.cc: Assign LOG_INFO priority to syslog calls
  + logrotate: Run as root/ceph
  + mon: OSDMonitor: Missing nearfull flag set
  + hobject: clarify is_snap and has_snapset for max
  + src/osd: relax the requirement that we scrub a whole hash value
  + osd_types,PG: force the map used in _scrub to be sorted correctly
  + PG: block writes and scan log for scrub using inclusive upper bound
  + rgw: RGWDataSyncCR fails on errors from RGWListBucketIndexesCR
  + rgw: add reinit/rebind logic (ldap)
  + rgw ldap: enforce simple_bind w/LDAPv3
  + rgw ldap:  protect rgw::from_base64 from non-base64 input
  + add ldap auth custom search filter feature patch - http://tracker.ceph.com/issues/17185
  + rgw_file: fix rename cases and unify unlink
  + rgw_file: refuse partial, out-of-order writes
  + rgw_file: implement rgw_setattr
  + rgw_file: refuse ops on deleted targets
  + rgw file: refuse to rename directories
  + rgw nfs: don't leak fh-&gt;mtx in lookup_handle()
  + rgw nfs: fix write continuations
  + rgw_file: unlock() must precede out label
  + rgw file: remove busy-wait in RGWLibFS::gc()
  + rgw_file: fix/update rgw_open flag arguments
  + rgw_file: fix/update rgw_create flag arguments
  + rgw_file: implement stateless write completion timer
  + rgw_file: add RGW_OPEN_FLAG_STATELESS as a synonym for V3
  + rgw_file:  log on write completion events at level 10
  + rgw file: allow RGWLibFS::write_completion_interval_s to be set in conf
  + librgw: add API version defines for librgw and rgw_file
  + ceph_timer: prefer using directive to typedef
  + rgw_file:  explain semantics of RGWFileHandle::commit
  + rgw_file: restore local definition of RGWLibFS gc interval
  + rgw: fix the wrong return value
  + rgw: rgw file fix bug of rgw_lookup can not exact match file name
  + rgw: remove duplicated calls to getattr
  + rgw_file: implement rgw_setattr
  + rgw_file: fix set_attrs operation
  + rgw_file:  pre-assign times
  + doc/radosgw: fix description of response elements 'Part'
  + librbd: new journal policy to disable initializing the journal
  + librbd: permit disabling journaling if in corrupt state
  + rgw: set args when rebuilding req_info
  + rgw: forward input data when forwarding set_bucket_version to master
  + init-radosgw: do not use systemd-run in sysvinit
  + messages: fix MMDSBeacon constructor
  + mon: don't crash on invalid standby_for_fscid
  + mds: make mds_role_t member order match constructor
  + mds: remove max_mds config option
  + os/filestore/FileJournal: fail out if FileJournal is not block device or regular file
  + cephx: Fix multiple segfaults due to attempts to encrypt or decrypt an empty secret and a null CryptoKeyHandler
  + common: only call crypto::init once per CephContext
  + mds: catch duplicates in DamageTable
  + client: fix segment fault in Client::_invalidate_kernel_dcache().
  + mds: remove fail-safe queueing replay request
  + client: fix readdir vs fragmentation race
  + ceph-create-keys: fix existing-but-different case
  + doc: clarify rbd size units
  + rbd: recognize lock_on_read option
  + rbd: fix up terminology in help texts
  + rbd: expose rbd unmap options
  + mds: s/used_ions/used_inos/
  + mds: log path with CDir damage messages
  + client: properly set inode number of created inode in replay request
  + librbd: ignore notify errors on missing image header
  + build: include more files in "make dist" tarball
  + cls/journal: add async client_update_state method
  + journal: allow to trim journal for "laggy" clients
  + librbd: optionally flag "laggy" journal clients disconnected
  + rbd: new command to disconnect journal client
  + rbd-mirror: decode_client_meta should return false on error
  + rbd-mirror: resync was possible only when image replayer start had succeeded
  + rbd-mirror: stop replay when client is disconnected
  + rbd-mirror: option to automatically resync after journal client disconnect
  + qa/workunits/rbd: demote/promote image on same cluster
  + rbd-mirror: include tag tid in bootstrap debug log messages
  + librbd: separate journal::TagData predecessor data into new struct
  + librbd: block RPC requests when demoting the image
  + librbd: normalize journal promotion/demotion events
  + rbd-mirror: demote/promote in same cluster results in split-brain
  + rbd-mirror: prevent syncing to non-primary image after failover
  + test: fixed unused function warnings in unittest_rbd_mirror
  + librbd: fix journal create compilation error
  + qa/workunits/rbd: before removing image make sure it is not bootstrapped
  + test/rbd: fix possible mock journal race conditions
  + qa/workunits/rbd: new mirroring forced promotion test
  + qa/workunits/rbd: fix remove mirrored image race conditions
  + journal: expose ability to retrieve partial tag list
  + librbd: unify journal event callbacks into single interface
  + librbd: helper class for quiescing in-flight async ops
  + librbd: new journal listener event for force promotion
  + journal: send update notification when tag allocated
  + journal: increase concurrency of journal recorder
  + journal: make librados call async in ObjectRecorder
  + rbd: Fix race between journal flush and append events
  + rbd-mirror: test: Fixed timeout problem in rbd_mirror_stress.sh
  + rbd-mirror: concurrent access of event might result in heap corruption
  + rbd-mirror: use pool id + global image id as deletion primary key
  + librbd: support deleting image by id instead of name
  + common: move AsyncOpTracker to common library
  + journal: fix AsyncOpTracker compiler error
  + rbd-mirror: wait for in-flight event commit notifications
  + journal: avoid holding lock while sending journal append
  + journal: use reverse iterator search to find flush record
  + journal: delay moving overflowed buffers until quiesced
  + journal: clean up object recorder closed/overflow callback
  + librbd: ignore attempts to request lock if already lock owner
  + rbd-mirror: potential race allocating tag during shut down
  + qa/workunits: reduce amount of client IO for rbd-mirror stress
  + journal: complete action only after notification completed
  + journal: ensure in-flight ops are complete destroying journaler
  + rgw: Let the default quota settings take effect during user creation
  + rgw: Do not bother to check against conf values during quota handling
  + rgw: Add documentation for RGW default quota
  + rbd: restrict mirror enable/disable actions on parents/clones
  + librbd: enable/disable of features is not allowed when already enabled/disabled
  + rbd: cleanup - Proxied operations shouldn't result in error messages if replayed
  + cls/rbd: set_features should adjust the mask to known features
  + librbd: update_features should handle Infernalis OSDs
  + librbd: potential deadlock closing image with in-flight readahead
  + librbd: block name prefix might overflow fixed size C-string
  + librbd: possible deadlock if cluster connection closed after image
  + librbd: corrected use-after-free in ImageWatcher
  + rgw:user email can modify to empty
  + rgw: remove user email index using helper function
  + rgw: don't fail if lost race when setting acls
  + rgw: fix versioned object IO error
  + doc: Radosgw multitenancy config paramater change
  + rgw:response information is error when geting token of swift account The header is X-Trans-Id instead of
    x-amz-request-id in the response header
  + doc: Modification for "TEST S3 ACCESS" section
  + Allow building RGW with LDAP disabled
  + tests: populate /dev/disk/by-partuuid for scsi_debug
  + AArch64: Detect crc32 extension support from assembler
  + rgw: mark configuration as converted to avoid corruption when running older admins
  + rgw: Fix missing master zone for default region conversion
  + rgw: avoid corruption when running old radosgw-admin on a newer rgw
  + mon/AuthMonitor: make AuthMonitor::IncType public
  + tools/ceph-objectstore-tool: add "update-mon-db" command
  + tools/ceph_monstore_tool: add "rebuild" command
  + doc: add rados/operations/disaster-recovery.rst
  + tools/rebuild_mondb: avoid unnecessary result code cast
  + tools/rebuild_mondb: kill compiling warning
  + tools/ceph_monstore_tool: update pgmap_meta also when rebuilding store.db
  + tools/ceph_monstore_tool: bail out if no caps found for a key
  + doc: fill keyring with caps before passing it to ceph-monstore-tool
  + tests: ceph-disk: force debug monc = 0
  + rgw: fix regression with handling double underscore
  + rgw: set correct instance on the object
  + rbd-nbd: fix kernel deadlock during teuthology testing
  + rbd-nbd: mask out-of-bounds IO errors caused by image shrink
  + librbd: invalidate cache before trimming image
  + librbd: ignore cache busy errors when shrinking an image
  + test: TestJournalReplay test cases need to wait for committed journal event
  + Don't loop forever when reading data from 0 sized segment.
  + journal: do not prematurely flag object recorder as closed
  + mon: fix missing osd metadata (again)
  + rgw: Do not archive metadata by default
  + ceph-disk: change ownership of init file to ceph:ceph
  + ceph-disk: Compatibility fixes for Python 3
  + ceph-disk: Use true integer division in get_dev_size
  + ceph-disk: Fix bug in FileLock
  + ceph-disk: Use context manager with FileLock
  + ceph-disk: Use os.path.isabs instead of .startswith('/')
  + ceph-disk: Set space_symlink to the path, not file object
  + ceph-disk: PEP8ify
  + ceph-disk: allow using a regular file as a journal
  + rgw: handle empty POST condition
  + librbd: discard after write can result in assertion failure
  + qa: remove EnumerateObjects from librados upgrade tests
  + rgw: fix put_acls for objects starting and ending with underscore
  + ceph-post-file: migrate to RSA SSH keys
  + ceph-post-file: Ignore keys offered by ssh-agent
Commits necessary to resolve merge conflicts
  + Revert "ceph_disk.main: s/get_dev_size/get_free_partition_size/"
  + build/ops: include man pages in tarball explicitly
Fix boo#1007216
  + Revert "Revert "Drop any systemd imposed process/thread limits""
Fix boo#1008501
  + ceph_volume_client: fix _recover_auth_meta() method
  + ceph_volume_client: check if volume metadata is empty
  + ceph_volume_client: fix partial auth recovery

- ibs_constraints.sh: bash script to update _constraints for IBS
- pre_checkin.sh: add ibs_constraints.sh to ceph-test Sources list

- Update to version 10.2.2+git.1475131634.4dc33de:
  + Revert "rpm: drop obsolete python-ceph-compat metapackage"
  + AArch64: Detect crc32 extension support from assembler (boo#999688)

- Update to version 10.2.2+git.1472752588.dada3f3:
  + spec: add ssl requirements
  + rgw: blanket apply NO_SSL_DL to civetweb sections
  (boo#990438)
</description>
  <summary>Security update for ceph</summary>
</patchinfo>

Places

File _patchinfo of Package patchinfo (Revision d5b70839882b509bdb71f7f75eee9138)

Places