summaryrefslogtreecommitdiffstats
path: root/src (follow)
Commit message (Collapse)AuthorAgeFilesLines
* osd: For testing full disks add injectfull socket commandDavid Zafman2017-04-172-1/+33
| | | | Signed-off-by: David Zafman <dzafman@redhat.com>
* common: Bump ratio for backfillfull from 85% to 90%David Zafman2017-04-171-1/+1
| | | | Signed-off-by: David Zafman <dzafman@redhat.com>
* common: Remove unused config option osd_recovery_threadsDavid Zafman2017-04-171-1/+0
| | | | Signed-off-by: David Zafman <dzafman@redhat.com>
* osd: too_full_for_backfill() returns ostream for reasonDavid Zafman2017-04-174-20/+14
| | | | Signed-off-by: David Zafman <dzafman@redhat.com>
* osd: Fail-safe full is a hard stop even for mdsDavid Zafman2017-04-171-1/+4
| | | | | | | We can't allow OSD to become non-startable even if mds could be writing as part of file removals. Signed-off-by: David Zafman <dzafman@redhat.com>
* test: Switch from pg to osd for set-*-ratio commandsDavid Zafman2017-04-171-15/+15
| | | | | | Testing of 6422e0a220fb3f32ccae50e0c7e52dc9984685c6 Signed-off-by: David Zafman <dzafman@redhat.com>
* osd: Remove unused argument to clear_queued_recoveryDavid Zafman2017-04-171-1/+1
| | | | Signed-off-by: David Zafman <dzafman@redhat.com>
* osd: Increase osd_backfill_retry_interval to 30 secondsDavid Zafman2017-04-171-1/+1
| | | | Signed-off-by: David Zafman <dzafman@redhat.com>
* ceph-objectstore-tool: cleanup commentDavid Zafman2017-04-171-1/+1
| | | | Signed-off-by: David Zafman <dzafman@redhat.com>
* osd: Fix log messageDavid Zafman2017-04-171-1/+1
| | | | Signed-off-by: David Zafman <dzafman@redhat.com>
* Merge pull request #12536 from ilc/cls_rgw_minor_refactorCasey Bodley2017-04-174-81/+143
|\ | | | | | | | | | | cls/rgw: Clean up the "magic string" usage in the cls layer for RGW. Reviewed-by: Casey Bodley <cbodley@redhat.com> Reviewed-by: Adam C. Emerson <aemerson@redhat.com>
| * cls/rgw: Clean up the "magic string" usage in the cls layer for RGW.Ira Cooper2016-12-164-81/+143
| | | | | | | | | | | | | | This set of changes, should make following what is going on across the cls barrier much easier for people using code browsing tools. Signed-off-by: Ira Cooper <ira@redhat.com>
* | Merge pull request #14046 from songbaisen/m7John Spray2017-04-172-11/+6
|\ \ | | | | | | | | | | | | mgr: optimization some judgment and adjust the debug remove value in register_new_pgs Reviewed-by: John Spray <john.spray@redhat.com>
| * | mgr: optimization some judgment and adjust the debug remove value in ↵songbaisen2017-04-172-11/+6
| | | | | | | | | | | | | | | | | | register_new_pgs Signed-off-by: song baisen <song.baisen@zte.com.cn>
* | | Merge pull request #14267 from yunfeiguan/masterJohn Spray2017-04-151-0/+1
|\ \ \ | | | | | | | | | | | | | | | | cephfs: fix mount point break off problem after mds switch occured Reviewed-by: Yan, Zheng <zyan@redhat.com>
| * | | cephfs: fix mount point break off problem after mds switch occuredYunfeiGuan2017-04-101-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The hot-standby become active as we expected but the mount piont broken strangely when the active mds is down. The root reason is the new mds use last_cap_renews decoded from ESesson::replay in find_idle_sessions and wrongly killed the session. Maybe we should reset session->last_cap_renew to the current time when server send OPEN to client in reconnect stage. Fixes: http://tracker.ceph.com/issues/19437 Signed-off-by: Guan yunfei <yunfei.guan@xtaotech.com>
* | | | Merge pull request #13587 from yanghonggang/masterJohn Spray2017-04-154-7/+60
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | cephfs: fix write_buf's _len overflow problem Reviewed-by: John Spray <john.spray@redhat.com>
| * | | | cephfs: fix write_buf's _len overflow problemYang Honggang2017-04-134-7/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After I have set about 400 64KB xattr kv pair to a file, mds is crashed. Every time I try to start mds, it will crash again. The root reason is write_buf._len overflowed when doing Journaler::append_entry(). This patch try to fix this problem through the following changes: 1. limit file/dir's xattr size 2. throttle journal entry append operations Fixes: http://tracker.ceph.com/issues/19033 Signed-off-by: Yang Honggang joseph.yang@xtaotech.com
* | | | | Merge pull request #13522 from singler/wip-reconnect-clientJohn Spray2017-04-152-2/+11
|\ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | client/Client.cc: after reset session from MDS - reconnect Reviewed-by: Yan, Zheng <zyan@redhat.com>
| * | | | | client/Client.cc: add feature to reconnect client after MDS resetHenrik Korkuc2017-04-122-2/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Client.cc marks session as stale instead of reconecting after received reset from MDS. On MDS side session is closed so MDS is ignoring cap renew. This adds option to reconnect stale client sessions instead of just marking sessions stale. Fixes: http://tracker.ceph.com/issues/18757 Signed-off-by: Henrik Korkuc <henrik@kirneh.eu>
* | | | | | Merge pull request #14207 from jcsp/wip-19401John Spray2017-04-151-0/+17
|\ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | mds: ignore ENOENT on writing backtrace Reviewed-by: Yan, Zheng <zyan@redhat.com>
| * | | | | | mds: ignore ENOENT on writing backtraceJohn Spray2017-04-131-0/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We get ENOENT when a pool doesn't exist. This can happen because we don't prevent people deleting former cephfs data pools whose files may not have had their metadata flushed yet. http://tracker.ceph.com/issues/19401 Signed-off-by: John Spray <john.spray@redhat.com>
* | | | | | | Merge pull request #14490 from liewegas/wip-no-luminousSage Weil2017-04-152-11/+43
|\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | mon: add mon_debug_no_require_luminous Reviewed-by: Kefu Chai <kchai@redhat.com>
| * | | | | | | mon/OSDMonitor: require_luminous_osds for pg-remap[-items]Sage Weil2017-04-141-0/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Sage Weil <sage@redhat.com>
| * | | | | | | mon/OSDMonitor: slightly better debug outputSage Weil2017-04-131-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Sage Weil <sage@redhat.com>
| * | | | | | | mon/OSDMonitor: fix creating pg trimming during upgradeSage Weil2017-04-131-5/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The creating pgs are in the PGMap with the CREATING state bit set. Signed-off-by: Sage Weil <sage@redhat.com>
| * | | | | | | mon/OSDMonitor: fix initial map when require_luminous_osds not set on mkfsSage Weil2017-04-131-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If we don't set the luminous flag, we should not set the new luninous fields or else we'll get a crc mismatch. (Funnily that happens in the epoch where the flag is eventually set and the encoded map finally includes the field we have set in memory.) Signed-off-by: Sage Weil <sage@redhat.com>
| * | | | | | | mon/OSDMonitor: ignore health checks about require_luminous if debug flag is setSage Weil2017-04-131-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This makes it tedious for teuthology health checks to proceed when we deliberately run luminous osds without this flag. Signed-off-by: Sage Weil <sage@redhat.com>
| * | | | | | | mon/OSDMonitor: add mon_debug_no_require_luminousSage Weil2017-04-132-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add debug option not to set the require_luminous osdmap flag on mkfs, for debugging purposes. Signed-off-by: Sage Weil <sage@redhat.com>
* | | | | | | | Merge pull request #14553 from guihecheng/rgw_file-revertMatt Benjamin2017-04-142-3/+10
|\ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | rgw_file: fix up potential race condition verified f23
| * | | | | | | | rgw_file: add locks to fix potential raceGui Hecheng2017-04-141-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Gui Hecheng <guihecheng@cmss.chinamobile.com>
| * | | | | | | | Revert "rgw_file.h: fix potential data race condition"Gui Hecheng2017-04-141-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 3f7b411e1e304e8a6f295dc50b8cf3c6c496f4cc. Causing hung up due to double lock. Signed-off-by: Gui Hecheng <guihecheng@cmss.chinamobile.com>
* | | | | | | | | Merge pull request #14546 from dachary/wip-19628-directory-bootKefu Chai2017-04-141-38/+63
|\ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ceph-disk: enable directory backed OSD at boot time Reviewed-by: Kefu Chai <kchai@redhat.com>
| * | | | | | | | | ceph-disk: enable directory backed OSD at boot timeLoic Dachary2017-04-141-38/+63
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | https://github.com/ceph/ceph/commit/539385b143feee3905dceaf7a8faaced42f2d3c6 introduced a regression preventing directory backed OSD from starting at boot time. For device backed OSD the boot sequence starts with ceph-disk@.service and proceeds to systemctl enable --runtime ceph-osd@.service where the --runtime ensure ceph-osd@12 is removed when the machine reboots so that it does not compete with the ceph-disk@/dev/sdb1 unit at boot time. However directory backed OSD solely rely on the ceph-osd@.service unit to start at boot time and will therefore fail to boot. The --runtime flag is selectively set for device backed OSD only. Fixes: http://tracker.ceph.com/issues/19628 Signed-off-by: Loic Dachary <loic@dachary.org>
* | | | | | | | | | Merge pull request #14551 from tchaikov/wip-19584Kefu Chai2017-04-141-0/+13
|\ \ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | mon/OSDMonitor: transit creating_pgs from pgmap when upgrading Reviewed-by: Sage Weil <sage@redhat.com>
| * | | | | | | | | | mon/OSDMonitor: transit creating_pgs from pgmap when upgradingKefu Chai2017-04-141-0/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | there could be some pg(s) still being created when we are upgrading to luminous, and the pools holding them are not changed in the sense of pg_pool_t::last_change after the upgrade and before we scan for creating pgs. in that case, the existing update_pending_creatings() will fail to collect the pgs being created before the upgrade. with this change, the creating_pgs in pgmap are also used for updating the OSDMonitor's creating_pgs if it's updated. but we should stopupdating the pgmap once the upgrade completes. i.e. stop dispatching MSG_PGSTATS messages to PGMonitor if the quorum and all osds are luminous. Fixes: http://tracker.ceph.com/issues/19584 Signed-off-by: Kefu Chai <kchai@redhat.com>
* | | | | | | | | | | Merge pull request #14435 from liewegas/wip-bluestore-deferredSage Weil2017-04-143-4/+27
|\ \ \ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | os/bluestore: tune deferred_batch_ops separately for hdd and ssd Reviewed-by: Igor Fedotov <ifedotov@mirantis.com>
| * | | | | | | | | | | os/bluestore: tune deferred_batch_ops separately for hdd and ssdSage Weil2017-04-133-4/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Sage Weil <sage@redhat.com>
* | | | | | | | | | | | Merge pull request #14489 from liewegas/wip-bluestore-fix-flushSage Weil2017-04-141-1/+4
|\ \ \ \ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | os/bluestore: fix narrow osr->flush() race
| * | | | | | | | | | | | os/bluestore: fix narrow osr->flush() raceSage Weil2017-04-121-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Sage Weil <sage@redhat.com>
* | | | | | | | | | | | | Merge pull request #13859 from jcsp/wip-19204John Spray2017-04-141-2/+4
|\ \ \ \ \ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | mds: shut down finisher before objecter Reviewed-by: Yan, Zheng <zyan@redhat.com>
| * | | | | | | | | | | | | mds: shut down finisher before objecterJohn Spray2017-04-131-2/+4
| | |_|_|_|_|_|/ / / / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Some of the finisher contexts would try to call into Objecter. We mostly are protected from this by mds_lock+the stopping flag, but at the Filer level there's no mds_lock, so in the case of file size probing we have a problem. Fixes: http://tracker.ceph.com/issues/19204 Signed-off-by: John Spray <john.spray@redhat.com>
* | | | | | | | | | | | | Merge pull request #14505 from jcsp/wip-19566John Spray2017-04-141-2/+4
|\ \ \ \ \ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | mds: fix mgrc shutdown Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
| * | | | | | | | | | | | | mds: fix mgrc shutdownJohn Spray2017-04-131-2/+4
| |/ / / / / / / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Were previously only tearing MgrClient down when not holding a rank, leading to it trying to continue to run after monclient was shut down. Fixes: http://tracker.ceph.com/issues/19566 Signed-off-by: John Spray <john.spray@redhat.com>
* | | | | | | | | | | | | Merge pull request #14398 from jcsp/wip-19551John Spray2017-04-141-9/+49
|\ \ \ \ \ \ \ \ \ \ \ \ \ | |_|_|_|/ / / / / / / / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | mon: emit cluster log messages on MDS health changes Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
| * | | | | | | | | | | | mon: emit cluster log messages on MDS health changesJohn Spray2017-04-131-9/+49
| |/ / / / / / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously, when we got a beacon that updated the health metrics for an MDS, the user would just see mysterious-looking cluster log messages indicating a rising fsmap epoch number. It would be good to do this for health messages in general at some point, but for now just do it for the MDS ones. Fixes: http://tracker.ceph.com/issues/19551 Signed-off-by: John Spray <john.spray@redhat.com>
* | | | | | | | | | | | Merge pull request #14469 from badone/wip-return-error-on-missing-client-idBrad Hubbard2017-04-141-5/+10
|\ \ \ \ \ \ \ \ \ \ \ \ | |_|_|/ / / / / / / / / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | mds: Return error message instead of asserting Reviewed-by: John Spray <john.spray@redhat.com>
| * | | | | | | | | | | mds: Return error message instead of assertingBrad Hubbard2017-04-131-5/+10
| | |_|_|/ / / / / / / | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For "session evict" admin socket command return an error message when we receive an invalid/missing client_id parameter rather than asserting. Signed-off-by: Brad Hubbard <bhubbard@redhat.com>
* | | | | | | | | | | Merge pull request #14420 from optimistyzy/329Sage Weil2017-04-131-146/+223
|\ \ \ \ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | os/bluestore/NVMEDevice: Add multiple thread support for SPDK I/O thread Reviewed-by: Haomai Wang <haomai@xsky.com>
| * | | | | | | | | | | bluestore, NVMEDevice: Add multiple thread support for SPDK I/O threadZiye Yang2017-04-121-146/+223
| |/ / / / / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously, we only have one thread to do SPDK I/O, and this patch adds the multiple thread support. In this first version, we use the tid of the thread to map the I/Os of this thread to the corresponding SPDK I/O thread. Signed-off-by: Ziye Yang <optimistyzy@gmail.com>
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-20 21:57:23 +0100