Gantt
The chart was truncated because it exceeds the maximum number of items that can be displayed (500)
Fix #5545: UI: Admin - Password Retrieval Feature #7070: Remove plugin system from the cthulhu Bug #8731: Fix SELinux for Calamari on RHEL7 Feature #9184: Calamari should use actual OSD devices on graph page Bug #9191: remove workaround for grains in relax_salt_perms.sls Feature #9275: Use FQDN for ceph_deploy AND ServerName in calamari.conf Bug #10223: asok handling code in minion module should ignore non-service things Bug #10496: mon counters incorrectly calculated when mon_status is warn Feature #6862: Gather and store historical cluster state Feature #6863: Expose historical cluster state in REST API Feature #6866: Backend support for account recovery by email Feature #6867: Security for graphite connections Feature #6876: Efficient latest-value querying for graphite stats Feature #6881: Diamond collector should use librados directly Feature #6898: Per-pool PG details and state counters Feature #6906: Manage pool snapshots Feature #7078: Respect GPG auth when installing salt+diamond Feature #7175: On-demand load of full PG detail Feature #7204: Internationalizable events and request descriptions Feature #7279: Proxy support in calamari bootstrap Feature #7410: Add some examples for bulk DELETE and PATCH to docs Fix #7419: 403 Errors on API return HTML Fix #7527: Document expected response codes in REST API Feature #7559: /api/v2/key entries attaching extra metadata to help UI Feature #7683: Collect ceph version per service. expose it through the API Feature #7811: Show a friendly message when apache is running but calamari-ctl isn't done Feature #7838: Metadata tags on UserRequests Feature #7839: Metadata tags on events Feature #7840: Bulk requests (many requests in one) Feature #7841: Give events machine-readable type identifier Feature #7842: Make UserRequests persistent Feature #7910: Consume salt presence events for server upness Feature #7911: Use public salt interfaces only Fix #7929: Avoid generating "OSD went down" events during a stopped cluster coming back online Feature #8064: Generate postgres account credentials at install time Fix #8074: Remove workaround for saltstack #11919 Fix #8075: Remove workaround for saltstack #11928 Fix #8144: Communication failure between salt-master and cthulhu Feature #8157: Annotations API Feature #8244: Epic: Cache Tier Pool CRUD Feature #8245: Epic: Erasure Code CRUD Feature #8247: Epic: RBD Management Fix #8460: Log spam: "Re-opening connection to salt-master" warnings when no minions are present Fix #8463: Cleaner handling of CRUSH maps with non-hostname-named hosts Feature #8477: Add cache tier object to cthulhu Feature #8478: Expose cache tier object CRUD in rest-api Fix #8545: Handle clusters where FSID isn't set in ceph.conf Feature #8663: CRUSH: Add CRUD to calamari api for crush_rules Feature #8756: Update django to fix known security flaws Feature #8812: RFE for calamari-ctl initialize: show Calamari URL/user Fix #8828: Remove log handling from cthulhu Feature #9164: sense SELinux and warn about it Cleanup #9330: Reserve blue text for clickable items Cleanup #9331: Link to product doc during setup Cleanup #9346: Terminology Cleanup Cleanup #9348: Buttons and tooltips Cleanup #9349: Drop down arrows for drops Cleanup #9350: Clean up Delete Pool warning dialog Feature #9457: CRUSH: Filter retrieval of crush nodes for children and depth Feature #9590: LOCALE: Allow the user to select the locale Feature #9591: LOCALE: Support localization for 'Manage' Feature #9722: CRUSH: Allow weight of bucket nodes be set using the crush_node endpoint Feature #9764: CRUSH: Report Not modified on the crush_node resource when no changes would be made Feature #9847: LOCALE: add placeholders to displayed text in dashboard Feature #9848: LOCALE: add placeholders to displayed text in manage app Feature #9849: LOCALE: create a non-sense language locale for visual indication that strings on the display are not properly translated Feature #9850: LOCALE: create howto documentation for new translations Feature #9861: LOCALE: Document that locale is chosen based on browser Accept-Language header. Tasks #9862: LOCALE: check that features in changelog for release X.Y.Z have not added text in the UI that is not internationalizable. Tasks #9879: LOCALE: Upgrade Angular JS from 1.2 to 1.3 before doing i18n updates Feature #9923: BE: resolve a host to crush node, decorate the host details with that Bug #10556: remove workaround in rest-api Validating-serializer for msgpack-python unable to pack SortedDictWithMetadata Cleanup #11402: rename "calamari-clients" packages to "romana" Bug #12082: calamari-server postinstall should use supervisor conf.d if possible Bug #12270: Host info pop-up window is empty in Calamari UI Bug #12292: CentOS: Calamari UI uses lsb_distrib_description but salt is sending lsb_distrib_id Bug #12375: calamari_common salt_wrapper sucks for diagnosability Bug #12416: The cthulhu should have the thead clear the database and log. Support #12521: Add Time Filtering to the events target. Bug #12560: RGW type in /server endpoint shows ambiguous "client" string Feature #12561: Include more health info in /mon targets. Feature #12686: Include all Addition and Removal of Cluster objects in events Bug #13029: calamari stats broken on infernalis Feature #13609: Ceph support Plan tasks or Timing task Function Bug #13648: rest-api: api/v1/space returned a history value, not the latest value in the whisper file Bug #13649: Diamond: One of my OSD has no performance data because diamond can not deal with the situation of a disk mounted to 2 folder which caused by an unsuccssful unmount operation occasionally Bug #13650: Diamond: when pull out the disk first, then insert the disk in a few minutes later, the disk's iops will be extremly high, higher than 10 minillion. Bug #13651: cthulhu: When i change a host's hostname, calamari can not update its server_state.hostname. Feature #13747: Add rbd and radosgw-admin support to /cli Post target Bug #14017: update pg_num of pool will return 500error, Bug #14179: Error 500 :: Internal server error ::: ubuntu 14.04 Trusty Bug #14396: Calamari dashboard :: can't connect to the cluster Bug #14440: Calamari Rest API: "api/v2/cluster/<fsid>/cli " throws 503 server error Feature #9421: installing Salt 2014.1.10 via calamari-boostrap on RHEL6.5 has a missing requirement for python-libcloud Fix #10588: ice-tools needs to install ice_setup as a package Feature #11083: EPIC: as an administration I want to receive notifications when OSDs are likely to fail Feature #11290: refine SMART error reporting to have levels of severity Feature #11291: SMART, support for other RAID controllers Feature #11292: expose events in the API when smart reports errors Feature #11293: decorate OSD API endpoint with smart status Feature #11366: ceph.py assumed no '-' in ceph cluster name Cleanup #11658: admin in romana is unused. remove it. Bug #12080: calamari crush location needs to deal with configurable cephx keyring location Bug #12127: calamari virtualenv aplies SWIG flag cerrosaswarn too broadly Feature #12522: Add RGW data to endpoints Bug #11379: UI reports "Cluster Updates Are Stale. The Cluster isn't updating Calamari. Please contact Administrator" when browser clock is skewed more than 15min Support #14437: dashboard widgets IOPs and Usage are blank Tasks #1418: set up a no-atomic-ops gitbuilder Feature #2010: mon: check for slow performing osds Feature #2108: track object states to inform error injection/testing Feature #2111: msgr workloads Feature #2127: Save kernel core dumps on all of our test machines Subtask #2239: install + configure package everywhere Feature #2283: The ceph command should time out Feature #2472: osd: add opaque 'class <name> <foo>' cap that class can interpret/enforce Feature #2578: rados ager Fix #3188: osd: close read hole Feature #3273: mon: simple dm-crypt key management Feature #3775: log: stop logging in statfs reports usage above some threshold Feature #4005: Add perftools to the kernel debian package script Fix #5388: osd: localized reads (from replicas) reordered wrt writes Bug #5634: auth startup reports "ObjectNotFound" when keyring file is unreadable Feature #5642: Show usable space via CLI/API Feature #5796: formatter: send output to ostream Feature #7291: EC: add mechanism for mon to detect and whitelist EC plugins which are globally available Cleanup #7586: deprecate rados_pool_create Documentation #7886: What's the policy on URL stability for public documentation? Fix #7890: erasure-code: last stripe is not truncated Feature #8033: Epic: Kerberos/LDAP Support Support #8462: related to integrity of objects Fix #8739: erasure-code: use LRU to cache decoding matrix in jerasure Bug #8741: osd: ec plugin leak Fix #8905: msgr: encode osd epoch in nonce to avoid misc OSD reconnect races Feature #8973: Add support for collecting usage information by namespace Feature #9034: erasure-code: better LRC strategy Bug #9109: ceph CLI: Help is missing -k keyring option Feature #9161: Cache warmup and ejection Feature #9222: annotate config options Fix #9371: ceph should fail when an argument is of the wrong type Fix #9601: erasure-code: ErasureCode::encode overhead is too high Feature #9647: osd: hard cap on PGs per OSD Support #9901: libgoogle-perftools4: tcmalloc performance regression on armhf Feature #9947: osd: store scrub error state in kv store; clear on peering event Feature #9948: osd: add scrub result query interface Feature #9949: librados: add ability to read a specific replica or shard Feature #9950: rados: add ability to read a specific replica/shard from CLI Feature #9951: librados, osd: per-object scrub operation Feature #9963: librados: improve get_objects and get_position interfaces Feature #9965: rados: new import from pipe/file Feature #9966: librados: set user_version operation Feature #10007: option to disable erasure code plugin version check Feature #10338: ceph: 'daemon list' command to enumerate daemon sockets we can connect to Feature #10386: add ceph-report-parsing Fix #10391: osd: an all/mostly delete pg log stalls during peering Subtask #10508: mon: finer-grain Paxos::is_readable() / PaxosService::is_readable() Fix #10789: mon: refactor OSDMonitor map methods to use MonSession*, not entity_inst_t Fix #10790: mon: move OSDMonitor osd_epoch value into MonSession Feature #11049: msgr: optionally encrypt messenger session Feature #11202: add stop_scrub command for ceph Bug #11298: aio gets EPERM when update-grub runs Feature #11299: generic plugin loader framework Fix #11377: ceph-disk: suppress non fatal partx errors Cleanup #11385: Standardize Ceph removal commands Bug #11389: should not "rm -rf /etc/ceph" when uninstalling ceph-common Feature #11668: erasure-code: measure bandwidth usage Documentation #11669: erasure-code: add cross datacenter example for lrc Tasks #11831: erasure-code: add chance_test_backfill_full to ceph-qa-suite tests Fix #11987: erasure-code: lrc k/m/l sanity check must apply to ruleset-failure-domain=osd Fix #12049: replace snprintf(s,...); x = strlen(s); with x = snprintf... Documentation #12052: document include/xlist.h Feature #12221: replica read: do proper locking on replica to avoid reading during an in-progress write Feature #12404: "ceph pool set-quota max_bytes" fails to work Feature #12430: libmailrados: Mailbox storage on RADOS Feature #12601: cmake Feature #12749: add an option for ceph_test_rados to have a very skewed distribution for accesses Fix #12953: mon: cull / scrub unused crush rules when pools are deleted Bug #12966: ceph-post-file key should only be readable by root Feature #13505: scrub/repair: persist scrub results. Fix #14362: should make librocksdb static link default in configure.ac Bug #14626: Ceph CLI + python bindings restrict string inputs to ASCII, but mon doesn't Documentation #15182: Write Developer Guide Bug #15289: src/test/cephtool-test-mon.sh Hooked allocator frame not found hides error (arm64) Support #17722: OSD`s doesnt start after reboot Bug #17724: SubsystemMap.h: 62: FAILED assert(sub < m_subsys.size()) Bug #18305: ceph-osd systemd unit files incomplete Cleanup #19654: crush: deprecate tree buckets Feature #19975: RFE: expose OP_OMAP_RMKEYRANGE in librados and cls Bug #20057: ceph-osd: some flags are not documented in the help output Fix #20224: ceph-disk deactivate fails with --verbose, and error message is confusing Cleanup #299: catch std::bad_alloc and die with helpful error in log on ENOMEM Tasks #2109: qa/benchmark: Explore using Filebench for benchmarks / stress testing Feature #2198: add an option to force a down osd to be marked immediately out Subtask #2201: Document old design Subtask #2203: implement new design Feature #2362: rados: support omap (leveldb) and locator key in import/export Tasks #2789: pkgconfig files for libraries Feature #2934: crush: create a visualizer for crush maps Feature #3080: rados: support asynchronous replication of some sort Feature #3249: Efficient access in distributed data centers Feature #3355: make Ceph build faster, take less RAM to compile Feature #3448: Minimize recovery time of a degraded cluster Feature #3511: figure out how to best set the heartbeat grace periods Feature #3805: log: detect dup messages Feature #4239: Automatically determine routing data based on host routing data (entirely deprecate public/cluster explicit specification) Feature #4278: add ability to make ceph ignore io errors for non-meta/internal data Feature #4516: qa: add test to compile a simple program against librados, librbd, and libcephfs Fix #4712: recovery from partially partitioned networks Bug #4733: Decode exception descriptions don't show function name Feature #4781: Journal entries should record last known committed entry Feature #4903: OpTracker: support hierarchies of request classes Feature #4986: create and automate scenario based testing Feature #5041: Deep scrub CPU limit behavior Feature #5042: Backport option to disable deep scrub to bobtail Tasks #5123: document src/objclass/objclass.h Feature #5141: Some clone errors aren't repaired Feature #5143: objclass: maintain global namespaces Feature #5148: repair should handle snapset/clone discrepancies Feature #5158: Objecter: support multi-read-from-replica Fix #5279: pipeline large object recovery Feature #5419: cephtool: sanitize extra args before configuring cluster handle Tasks #6185: expand upgrade tests to be able to test downgrade Feature #6198: packaging, admin_socket: create ceph group, make socket be group writable Feature #6221: Objecter,OSD: make it easy to determine when the network is misbehaving Feature #6227: make osd crush placement on startup handle multiple trees (e.g., ssd + sas) Fix #6527: backfill_toofull => try backfilling another replica Feature #6568: ceph-rest-api authentication Feature #6687: Ability to set up/down/in/out based on CRUSH hierarchy Feature #6759: Allow partial dump of pg statistics Cleanup #6766: keyring leading spaces Feature #6924: Add a new Info state to ceph health Feature #7104: rest-api: support commands requiring 'w' cap without 'rw' cap Feature #7108: build: error out on inappropriate use of dout() Feature #7114: Hinted recovery Feature #7196: qa: test encoding semantics, not just being able to decode/encode different versions Feature #7287: Make units of measurement constant and definable in ceph commands Cleanup #7314: remove dead code from OSDMonitor::should_propose Feature #7347: Per CRUSH type mon_osd_down_out_interval Bug #7407: Illogical defaults in Python bindings: conffile=None loads no config, conffile='' loads config from default path Feature #7484: better wording when authentication fails Feature #7700: Create a health severity between OK and WARN Fix #7711: OpTracker output doesn't include op size for subops Feature #7723: Cancel RADOS Bench writes and still do reads Tasks #7864: please clarify copyright and the license Feature #7884: investigate having the messenger (or dispatch q?) in the osd limit the number/cost of outstanding client IO requests Feature #8189: ceph: display tier relationships visually Feature #8195: shorten window of highest risk during recovery Feature #8348: include "ChangeLog" and/or "NEWS" files to release tarball Feature #8546: add nonce/sequence number associating the several pg metadata keys which are always updated atomically Cleanup #8590: better error reporting from ceph-disk Tasks #8637: leveldb: initial inquiries on how we can improve leveldb Feature #8965: Improve threading for ObjectCacher Feature #8992: Uniqueness between two or more CRUSH ruleset choose statements Feature #9097: request for tools/commands to see hits/misses on cache pools Feature #9305: erasure-code: cache minimum-to-decode results Feature #9306: erasure-code: optimize IsReadablePredicate to check data chunks instead of just rec_pred Feature #9383: admin socket: allow multiple pending requests Feature #9410: Crush optimization for unbalanced data/pg distribution Feature #9431: mon: audit read-only commands on a lower level than default Fix #9484: OSD: block until we have the same map as the client on pg commands Fix #9566: osd: prioritize recovery of OSDs with most work to do Feature #9594: stop backfill when osd becomes too full Cleanup #9600: rework bufferlist::*aligned* functions Fix #9689: ceph df reports % of global size used instead of MAX AVAIL 0.80.6 Cleanup #9756: Issues found by Clang Fix #9767: do not leak ceph-disk activate lock to the OSD Feature #9776: try to make address sanitizer work Fix #9778: forbid erasure code profile modifications that can modify data encoding Feature #9792: make it harder to remove cluster data pools Feature #9802: When replaced a disk, the CRUSH weight of the related host changed Cleanup #9930: gtest: update, move to submodule Subtask #9931: create selinux policies for ceph-mon, ceph-osd, ceph-mds Feature #9943: osd: mark pg and use replica on EIO from client read Support #10024: Cluster unreachable after restart Feature #10044: ECUtil::HashInfoRef should have a NONE value Feature #10054: OSD level metadata mismatch handling Feature #10055: PG metadata corruption detection and handling Feature #10056: Object metadata mismatch detection and handling Cleanup #10094: Create new git repo for json_spirit Cleanup #10110: librados: mark old objects_begin interface deprecated Fix #10244: double resource for setting up ceph-deploy Feature #10254: mon,osd: long-term non-clean PGs prevent osdmap trimming Feature #10300: Per-pool PG recovery prioritisation Feature #10457: Permanently mark daemons and clusters which use experimental features Support #10486: OSD Keeps Going Down Feature #10742: Initial pg_num choice is static, different from other default setting Feature #10759: rados command line gives no access to striper API Bug #10763: OSDs get marked as down in docker Feature #10786: cache pool: try not to keep around very old whiteout objects even if there is no cache pressure Feature #10796: Schedule scrubbing by considering PG's last_scrub_timestamp globally Feature #10865: Handle delete log entries in merge_log out of band without blocking peering Feature #10866: replicas need to track unstable objects to properly support replica reads Fix #10877: CLI error numbers are not described anywhere Feature #10917: mon add check to verify compatible version when joining cluster Feature #10941: Generic support for plugins installation and upgrade Fix #10943: invalid syntax in Apache config file examples Fix #10991: docs: federated rgw setup needs to be trimmed Fix #10992: docs: federated rgw setup keyring section should come earlier Fix #10993: docs: federated rgw setup create-pools section should specify what a user should create Fix #10994: docs: federated rgw setup create pools section should explain what ruleset-number is Fix #10995: docs: federated rgw setup create pools section should explain the need for dotted names Fix #11002: docs: federated rgw setup uses Debian-based locations/commands Fix #11003: docs: federated rgw setup assumes that the user has setup an RGW instance already Fix #11004: docs: federated rgw setup doesn't warn about FastCgiWrapper Fix #11005: docs: federated rgw setup needs to normalize variable names in example configs Fix #11006: docs: federated rgw setup assumes user is proficient in ceph-deploy Fix #11007: docs: federated rgw setup is not explicit about short hostname needs Fix #11008: docs: federated rgw setup should post example commands for confirmation Feature #11017: Improve scrubbing throutput Feature #11024: throttle objects and bytes / second for scrub/deepscrub Feature #11025: ceph health warn if pgs are not being scrubbed quickly enough for the scrub intervals Feature #11028: LMDB key/value backend for Ceph Feature #11094: New feature : Ceph should warn if the kernel is not turned for CEPH OSD daemons Feature #11109: move ceph-deploy(8) man page into ceph-deploy package Fix #11116: ceph-disk breaks when using non-existing disks Feature #11207: mons: add switch to never delete pools, only rename (pool foo->deleted_foo?). Feature #11240: make sure notcmalloc builds continue working Feature #11268: erasure-code: hitchhiker plugin Feature #11362: test: verify that expected feature bits are present in a cluster Feature #11403: add a command to the ceph tool which returns whether there are pending pg removals on any osd Feature #11443: Elector: throttle election attempts from DoSing peers Feature #11521: Certain configuration options not safe for production builds Feature #11525: improve OPTION macro to add a check function for invalid options. Feature #11544: add HELP command to ceph interactive shell Bug #11580: pg a copy_from->pg b peered (blocked on backfill blocked on local reservation)->pg a backfill holding local reservation blocked on copy_from blocked obc Feature #11600: OSD: ensure that filestore queue_transaction never blocks Feature #11631: msgr: throttle max open sockets Feature #11633: mon: automatically blacklist clients after failed auth attempts Feature #11689: change ceph-mon TCP port to an IANA-registered port Fix #11820: Show osd as NONE in "ceph osd map <pool> <object> " output Feature #11823: Workqueue timeouts and the filestore sync_entry timeout should log something descriptive at derr log level when exceeded Feature #11837: mon: add ability to gradually remove an osd Tasks #11847: OSD crashes under cached cluster benchmark Feature #11904: ceph-cli: let "ceph report" support compressed json dump Feature #11943: erasure code plugin for tornado code Support #12085: data is no well-distributed among osds of host with staw algorithm Cleanup #12126: Does the ceph debian package really need binutils? Support #12177: removing cached objects doesn't quickly remove tiered ec objects Feature #12233: init scripts: try to detect when a daemon named in ceph.conf is duplicating one whose folder is "managed" Tasks #12260: Messenger: use setsocketopt to set sk priority field based on message priority prior to sending each message Feature #12403: Implement a minimal access keyring for read-only access on a Ceph cluster Feature #12406: Support for building on Alpine Linux Bug #12431: Please publish rados-java artifacts to central maven repository, unable to load Rados Library Support #12468: Why copy a 3T image without any data takes 99 minutes? Feature #12514: PGMonitor::get_health get ops' blocked the detail information, show too much Support #12760: FileSystem become read-only after over 6 hour massive read and write Tasks #12797: create the upgrade test suite for gmt and sortbitmap change Feature #12864: Make "rados bench" support json output Feature #12915: Translate fadvise flag NOCACHE into DONTNEED if first access object Cleanup #12961: OSDMonitor refactor Fix #13005: compile warning on gcc-4.8.3 Feature #13073: automatically set journal file name as osd.{osd-num} with ceph-deploy. Feature #13084: ceph-disk: will not work with an software array (md) as journal destination Feature #13120: osd: prioritize more degraded PGs for recovery by considering the missing_loc of the PG Feature #13163: Add flag to delete a pool and its cache tier. Cleanup #13179: ceph-disk: use systemd-escape instead of hardcoded translation Bug #13312: ceph-authtool(8) doesn't mention namespace Feature #13319: ceph-objectstore-tool: add feature to adjust object_info size Feature #13358: OSD assert with 'Current_Pending_Sector' Feature #13419: On osd startup we should log all configuration settings Feature #13459: librados: c + python api for list_snaps() Feature #13506: scrub/repair: add librados APIs Feature #13509: add checksum for the decode/encode Feature #13521: AsyncMessenger: Support memory pool per worker Tasks #13617: Async support for Ceph-fio-objectstore Tasks #13618: objectstore_bench did not support newstore Feature #13641: osd assert is too crudely, maybe notify mon before quit is better. Cleanup #13941: Create can_discard_op_initial_decode and can_discard_op_final_decode to distinguish tests which require the op to be fully decoded Support #13951: ceph-disk list not support ID_PART_ENTRY_* fields Feature #14031: EC overwrites Feature #14033: Activation support for EC Overwrites Feature #14035: EC overwrites -- cleanup of ReplicatedPG::do_op to actually allow the new writes Feature #14036: EC overwrites: PGBackend needs to be replumbed to support overwrites Feature #14037: PGBackend: add a way for the Backend to notify ReplicatedPG that a version has been applied Feature #14038: ECBackend: delay writes which cause a clone to be created on an object with unapplied writes while those applies are forced Feature #14039: ECBackend cache extents with unapplied writes Feature #15621: EC Overwrites: create temporary testing pool flag for enabling ec supporting features Feature #15623: EC Overwrites: ECBackend TPC support Feature #17156: update ec overwrites implementation to support truncate Feature #17157: update ec overwrites implementation to support zero Feature #14032: EC Overwrites: Phase 2: Peering support for ec overwrites Feature #14034: EC Overwrites: Phase 2: PGLog support for recording participants Feature #15622: performance counters for ECbackend cache Feature #17158: EC Overwrites: work out implications of recovery below min_size pushing the can_rollback_to line forward Feature #17500: updated ec overwrites to support fadvise Feature #17501: ec overwrites write zeroes efficiently Feature #17651: EC Overwrites: Missing Test Case: aligned writes Feature #17652: EC Overwrites: Missing Test Case: multiple writes, single operation Feature #17653: EC Overwrites: Missing Test Case: zero Fix #14083: degrade a log message to a default level Bug #14093: scripts/ceph-release-notes does not handle merge description Feature #14104: add backfill targets to osd stats and a way to dump all osd stats to json Feature #14183: Show the totall pg on each osd. Cleanup #14202: the method 'Message::decode_message' is too long Feature #14216: Show the pool quota info on ceph df detial command Cleanup #14334: Crush when get immediate parent quickly reply. Feature #14346: Command to override Incomplete PGs Feature #13508: scrub/repair: repair corrupted/missing objects Bug #14366: unsafe handle_conf_change() methods Feature #14416: Write software version to data store after successful start Documentation #14439: ceph-crush-location manpage missing Support #14448: Installation (Quick) mount disk question Feature #14454: mon: report pool deletion in "ceph -w" Support #14588: Can't add new osd an existing cluster. Cleanup #14833: move rados-classes from /usr/lib to /usr/lib/ceph Feature #14852: Make SOCKET_BACKLOG configurable Fix #14932: Improve ceph-disk error message when there is not enough space to create journal partition Feature #15157: osd: try to flush/evict object from cache tier based on time period Bug #15203: NEON error when building ceph on ARM v7 targets Bug #15385: uniquely identify /tmp build directories Feature #15420: Enhance KeyValueDB to include atomic operator merge capability. Feature #15433: CAS Pools Feature #15434: Tiering: Object Redirects Feature #15751: Faster Peering -- primary remember infos/missings across interval if the primary doesn't change Feature #15788: Add --wait-for-complete flag to ceph pg scrub (and repair, and deep-scrub) Feature #15790: general pg inconsistency report Feature #15873: choose randomly among 3 different filestore_merge_threshhold/filestore_split_multiple values on osd startup in the qa suite Feature #15878: modify ceph-qa-suite to use tc to inject delays and resets for ceph daemon connections Support #15914: unit files missing after upgrade 10.2.1 Feature #16018: Use power crc32-vpmsum to accelerate crc32c performance on power8 platform Documentation #16021: OS Recommendations do not handle distribution-included packages well Bug #16131: messenger only uses IPv4 address of hostname Feature #16170: print out error on failure to load OSDMap in OSD::init -- EPERM and ENOENT should get specific messages explaining what happened Tasks #16174: SMR Benchmarking Scripts and libzbc code merge Feature #16180: AIO version of selfmanaged_snap_create/selfmanaged_snap_remove Feature #16261: ceph-objectstore-tool shouldn't fully lock the osd when performing read-only operations Support #16281: Lost xattr from all osd's Feature #16310: Take pipe::connect() returned errno to rados_connect() Part2 of http://tracker.ceph.com/issues/15308 Feature #16361: usability: allow decreasing pg_num when pg_num > pgp_num Feature #16462: Include more detail in perf dump Feature #16562: rados put: use the FULL_TRY flag to report errors when cluster is full Feature #16566: rados CLI's "rmomapkey" and "setomapval" should support binary keys Feature #16681: Creating EC pool using replicated ruleset should perhaps require a force flag Fix #16746: Fix the Ceph and Ceph UT compiling error in AArm64 Feature #16758: HEALTH_ERR status is misleading Bug #16786: Compiling with --without-cython seems to produce an entirely broken Ceph CLI, but the Ceph CLI still gets installed. Feature #16885: ceph should warn the administrator when it fails to open log_file for writing Fix #16912: Jewel ceph-dbg packages are missing for Debian-based distributions Feature #17071: Can't get pool's cache mode parameter via "ceph osd pool get" Feature #17166: Accelerate crush_do_rule with SIMD Fix #17433: [systemd]: Ceph doesn`t remove symlinks for systemd services Cleanup #17535: using RPC to eliminate complexity of network Feature #17597: leader should recognize its own clock skew instead of assuming the whole cluster is skewed Feature #17700: osd: add an interface to allow an object class to generate a warning in the central log Feature #17714: reweight-by-utilization needs a subtree option Tasks #17759: test radosgw with messenger failure injection and very short logs (osd min pg log entries = 10, osd max pg log entries = 20) Feature #17761: Support creating objects without network traffic Feature #17796: osdmap calculation frequency option request Support #17892: frequent blocked request impacting Ceph client IO Feature #18052: Replace past_intervals with more compact structure Feature #18160: support shared ccache dirs across users Feature #18167: ceph df should show correct pool usage Feature #18168: readline isn't configured or installed in ceph cli Feature #18267: ceph-disk may use blivet Feature #18394: Add S3 lifecycle configuration validate Feature #18400: Have OSD mark itself down while still running Feature #18438: Configurable OSD Heartbeat packet size (MTU) Feature #18486: RGW: Create users with a tenant via the admin ops api Support #18506: Where physically stored http headers for the location of the file in the ceph? Feature #18541: Support lifecycle validate in s3test Bug #18745: Python librados OmapIterator segfaults if used before op Feature #18851: Ability to add comments in certain views of Ceph daemons or status Feature #18966: Add mon_osd_down_out_subtree_max_osd Feature #18978: ceph-disk: Implement easy replace of OSD Documentation #19238: librados python binding reference is not autogenerated Cleanup #19284: Script to find broken URLs in Ceph source code. Feature #19310: Output only the latest require_{release}_osds flag Feature #19381: ceph_objectstore_tool,rados tool: add commands to both for exporting and importing a specified object to and from a common format Feature #19441: Merge snapshot of the cloned image back to the parent image Support #19548: 'journaling' feature affects write bandwidth Feature #19703: If there is one FULL OSD, All IO the the cluster will be halted(read only). Feature #19799: Expose snap trim speed along with client and recovery IO rates Tasks #19973: detect broken links in documentation automatically Bug #20233: Cannot remove striped object Feature #20300: run check_commands.sh as part of make check Fix #20305: ceph osd new -i should read from stdin Fix #20306: ceph-disk should use the python random function Feature #20311: mon/test: increase coverage of test-mon-msg Feature #20312: give scrub a start_time property Feature #20324: Change default filestore 'omap' backend to 'rocksdb' from 'leveldb' Support #20391: cluster health is ok, but client IO hang with 100% utils of ssd disk Feature #20553: Devuan support Cleanup #20685: Remove ceph_osd_mkfs function from src/ceph-disk/ceph_disk/main.py Fix #20695: http://www.irchelp.org/irchelp/new2irc.html [File not found] Cleanup #20698: Constants in code should be replaced with MACRO Feature #20768: Add API to cancel AIO requiest (for RBD and for Rados). Both for C and for Python. Feature #20786: ceph df should report %USED and MAX AVAIL taking quota into account Fix #20868: optimize some timer Cleanup #21075: Creating type alias for complex type definitions Support #21589: ceph 12.2.0 health check failed osd down Feature #21703: discover ip node in ceph cluster Bug #21747: Ceph ISCSI Gateway: required rpm packages not available for CentOS 7.4 Feature #21824: RDMA Direct Memory Operations Support #21913: Ceph cluster capacity much small than it should be Support #20994: 64pgs are in active+remapped state after initial deployment following official ceph deployment guide
|
« Previous | Next » |