summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* doc: Edited Joao's procedure for map consistency and changing IP addresses.wip-mon-docsJohn Wilkins2012-12-211-89/+69
| | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
* doc: add-or-rm-mons.rst: Add 'Changing Monitor's IPs' sectionJoao Eduardo Luis2012-12-201-1/+187
| | | | Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
* doc: add-or-rm-mons.rst: Clarify what the monitor name/id is.Joao Eduardo Luis2012-12-191-11/+18
| | | | Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
* OSDMap::dump: tag pg_temp mappings with pgidSamuel Just2012-12-181-0/+3
| | | | | Signed-off-by: Samuel Just <sam.just@inktank.com> Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
* Merge remote-tracking branch 'gh/testing' into nextSage Weil2012-12-172-1/+7
|\
| * v0.55.1v0.55.1Gary Lowell2012-12-122-1/+7
| |
* | Revert "objecter: don't use new tid when retrying notifies"Sage Weil2012-12-172-21/+3
| | | | | | | | | | | | | | | | | | This reverts commit c3107009f66bc06b5e14c465142e14120f9a4412. This appears to be causing problems in the objecter by corrupting the stack. Until that is resolved, let's revert. Signed-off-by: Sage Weil <sage@inktank.com>
* | mon: OSDMonitor: add option 'mon_max_pool_pg_num' and limit 'pg_num' accordinglyJoao Eduardo Luis2012-12-172-3/+5
| | | | | | | | | | | | | | | | | | Instead of having a hardcoded default, use a configurable one. It is limited to 65536 until future testing guarantees there is no side-effects of increasing it past this value, but by being adjustable the user still has the freedom to specify whatever maximum value he wants. Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
* | osdc/Objecter: prevent pool dne check from invalidating scan_requests iteratorSage Weil2012-12-152-9/+8
| | | | | | | | | | | | | | | | | | | | | | | | We iterate over ops and, if the pool dne and other conditions are true, we will immediately return ENOENT and cancel an op. Increment the iterator at the top of the loop to avoid invalidating it. We also need to switch to a map<>, because hash_map<> mutations may invalidate any/all iterators. Fixes: #3613 Signed-off-by: Sage Weil <sage@inktank.com>
* | qa: add a workunit for fsync-testerGreg Farnum2012-12-141-0/+8
| | | | | | | | | | | | | | | | | | | | It turns out that our suites don't exercise fsync, at least not very much (I couldn't find it in all the places I looked for it). This tester was written by Ted T'so and updated by Chris Mason; I just made it work on a smaller dataset (256MB) because 8GB against a small cluster takes more time than we want to wait. Signed-off-by: Greg Farnum <greg@inktank.com>
* | test: remove underscores from cephfs test namesNoah Watkins2012-12-142-15/+15
| | | | | | | | | | | | | | | | | | | | Google Test documentation strongly suggests avoiding underscores from unit test names to avoid accidental conflicts with their macro naming scheme. http://code.google.com/p/googletest/wiki/FAQ#Why_should_not_test_case_names_and_test_names_contain_underscore Signed-off-by: Noah Watkins <noahwatkins@gmail.com>
* | Merge branch 'wip_watch' into nextSage Weil2012-12-144-17/+143
|\ \
| * | OSD: put connection in disconnect_session_watches as well as the sessionSamuel Just2012-12-131-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | obc->watchers now has a ref to the connection as well. This piece of disconnect_session_watchers essentially parallels remove_watcher and should generally do the same thing. Signed-off-by: Samuel Just <sam.just@inktank.com>
| * | OSD: disconnect_session_watches obc might not be valid after we relockSamuel Just2012-12-131-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If disconnect_session_watches races with watch removal, the session might no longer have a valid obc ref. In that case, move on to the next obc. Note, there is no danger of any obcs being *added* to the session since the session/connection at this point is dead. Signed-off-by: Samuel Just <sam.just@inktank.com>
| * | OSD: pg might be removed during disconnect_session_watchesSamuel Just2012-12-121-1/+6
| | | | | | | | | | | | | | | | | | | | | We don't hold the osd_lock between the session->watches traversal and the obc checks. Signed-off-by: Samuel Just <sam.just@inktank.com>
| * | PG,ReplicatedPG: handle_watch_timeout must not write during scrub/degradedSamuel Just2012-12-123-2/+93
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, handle_watch_timeout will gladly write to an object while that object is degraded or is being scrubbed. Now, we queue a callback to be called on scrub completion or finish_degraded_object to recall handle_watch_timeout. The callback mechanism assumes that the registered callbacks assume they will be called with the pg lock -- and no other locks -- already held. The callback will release the obc and pg refs unconditionally. Thus, we need to replace the unconnected_watchers pointer with NULL to ensure that unregister_unconnected_watcher fails to cancel the event and does not release the resources a second time. Signed-off-by: Samuel Just <sam.just@inktank.com>
| * | ReplicatedPG:, remove_notify, put session after conSamuel Just2012-12-121-1/+1
| | | | | | | | | | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
| * | ReplicatedPG: only put if we cancel evt in unregister_unconnected_watcherSamuel Just2012-12-121-3/+7
| | | | | | | | | | | | | | | | | | | | | If we fail to cancel the callback, the callback will fire and release those resources. Signed-off-by: Samuel Just <sam.just@inktank.com>
| * | ReplicatedPG: watchers must grab Connection ref as wellSamuel Just2012-12-121-1/+7
| | | | | | | | | | | | | | | | | | | | | | | | Session refs are not really valid on their own, the corresponding Connection must remain live for at least as long as the Session. Signed-off-by: Samuel Just <sam.just@inktank.com>
| * | PG,ReplicatedPG: move write_blocked_by_scrub logic into a helperSamuel Just2012-12-112-9/+20
| | | | | | | | | | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* | | lockdep: Decrease lockdep backtrace skip by 1Sam Lang2012-12-141-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Skipping the top 4 (it starts at 0) calls in the backtrace actually skips the call that does the lock. Skip 3 instead. Signed-off-by: Sam Lang <sam.lang@inktank.com>
* | | mkcephfs: fix == -> =Sage Weil2012-12-141-1/+1
| | | | | | | | | | | | | | | | | | Another bashism. Signed-off-by: Sage Weil <sage@inktank.com>
* | | map-unmap.sh: use udevadm settle for synchronizationAlex Elder2012-12-141-16/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This script was heuristically using short sleep commands in order to give udev activity time to complete. There's a command "udevadm settle" which actually looks at the udev queue and waits until its processing is done. Much, much better. This rearranges the get_id function a bit too, breaking it into one function that gets the id and another that loops back and tries again after a short delay in the event the get_id fails. Signed-off-by: Alex Elder <elder@inktank.com>
* | | Merge branch 'wip-upstart' into nextSage Weil2012-12-144-5/+9
|\ \ \ | | | | | | | | | | | | Reviewed-by: Greg Farnum <greg@inktank.com>
| * | | ceph-disk-activate: mark dir as upstart-managedSage Weil2012-12-141-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Mark the directory so that upstart will manage the daemon. Eventually, this should be generalized to allow ceph-disk-* usage with other init systems. Signed-off-by: Sage Weil <sage@inktank.com>
| * | | upstart: make starter jobs consistentSage Weil2012-12-142-2/+1
| | | | | | | | | | | | | | | | Signed-off-by: Sage Weil <sage@inktank.com>
| * | | upstart: only start when 'upstart' file exists in daemon dirSage Weil2012-12-143-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We need to distinguish between daemons managed by upstart and sysvinit (and, eventually, systemd). Only start daemons when 'upstart' is present. Note that sysvinit will only start daemons when the 'host = ...' line is in ceph.conf, so there is a similar "opt-in". Signed-off-by: Sage Weil <sage@inktank.com>
* | | | ReplicatedPG: use default priority for Backfill messagesSamuel Just2012-12-141-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Backfill messages modify the stats on the replica and therefore must be sent with the same priority as sub_op_modify to ensure ordering. Using recovery_op_priority caused the following sequence: 1) Primary(1) sends MOSDPGBackfill FINISH with updated stats (v1) 2) Primary(1) sends SubOp modify for new client op with stats (v2) 3) Replica(2) receives SubOp with stats (v2) 4) Replica(2) receives MOSDPGBackfill FINISH with stats (v1) 5) Replica(2) responds and Primary(1) resets pgtemp making Replica(2) Primary(2) 6) PG stats on Primary(2) several ops old. Signed-off-by: Samuel Just <sam.just@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
* | | | ReplicatedPG: do not use priority from client opSamuel Just2012-12-141-2/+2
|/ / / | | | | | | | | | | | | | | | | | | | | | | | | There are internal ordering requirements which may be sensitive to assigned priority. We don't want a mix of priorities from old clients with priorities from new clients causing trouble. Signed-off-by: Samuel Just <sam.just@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
* | | Merge branch 'wip-3610' into nextSam Lang2012-12-1419-200/+1689
|\ \ \
| * | | client: Add config option to inject sleep for tickSam Lang2012-12-133-48/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Testing the tick delay with a fork/suspend is causing corruption in the lockdep code. This approach uses a config option to sleep the tick thread for a number of seconds, avoiding the entire fork/suspend mess. Signed-off-by: Sam Lang <sam.lang@inktank.com>
| * | | Merge remote branch 'origin/next'Josh Durgin2012-12-1313-20/+175
| |\ \ \
| * | | | filestore: Don't keep checking for syncfs if foundSam Lang2012-12-121-4/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Valgrind outputs a warning for unrecognized system calls, and does so for the syscall(__SYS_syncfs,...) and syscall(__NR_syncfs, ...) calls. This patch avoids making those calls (and the warning, when run in valgrind) if the syncfs libc call is available. INFO:teuthology.task.ceph.osd.1.err:--10568-- WARNING: unhandled syscall: 306 INFO:teuthology.task.ceph.osd.1.err:--10568-- You may be able to write your own handler. INFO:teuthology.task.ceph.osd.1.err:--10568-- Read the file README_MISSING_SYSCALL_OR_IOCTL. INFO:teuthology.task.ceph.osd.1.err:--10568-- Nevertheless we consider this a bug. Please report INFO:teuthology.task.ceph.osd.1.err:--10568-- it at http://valgrind.org/support/bug_reports.html. Signed-off-by: Sam Lang <sam.lang@inktank.com>
| * | | | doc: Updated per comments in the mailing list.John Wilkins2012-12-121-3/+2
| | | | | | | | | | | | | | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
| * | | | qa: exclude some more xfstestsJosh Durgin2012-12-111-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | These worked on a newer kernel, but I forgot I had not updated it for the final image. Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
| * | | | Merge branch 'next'Josh Durgin2012-12-113-5/+29
| |\ \ \ \
| * \ \ \ \ Merge branch 'next'Sage Weil2012-12-1134-139/+1464
| |\ \ \ \ \
| * | | | | | doc: document swift compatibilityYehuda Sadeh2012-12-113-1/+75
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add a table that specifies swift features compatibility Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * | | | | | docs: add rgw POST object as supported featureYehuda Sadeh2012-12-111-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * | | | | | Merge remote-tracking branch 'gh/next'Sage Weil2012-12-1035-146/+582
| |\ \ \ \ \ \
| * | | | | | | doc: fixed indent in python example.John Wilkins2012-12-101-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
| * | | | | | | doc: Added sudo to ceph -k command.John Wilkins2012-12-101-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
| * | | | | | | doc: Fixed typo.John Wilkins2012-12-101-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
| * | | | | | | Makefile.am: add missing flags to some tests targetsYehuda Sadeh2012-12-101-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | adding CRYPTO_CXXFLAGS to some targets. This is required when building --with-nss. Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * | | | | | | rgw: document admin api web interface.caleb miles2012-12-081-3/+1504
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: caleb miles <caleb.miles@inktank.com>
| * | | | | | | Merge branch 'next'Dan Mick2012-12-071-2/+4
| |\ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Merge of wip-rbd-export-progress Reviewed-by: Josh Durgin <josh.durgin@inktank.com>
| * \ \ \ \ \ \ \ Merge branch 'master' of https://github.com/ceph/cephJohn Wilkins2012-12-0722-104/+196
| |\ \ \ \ \ \ \ \
| | * \ \ \ \ \ \ \ Merge branch 'next'Josh Durgin2012-12-0722-104/+196
| | |\ \ \ \ \ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Merge of wip-rbd-create: Reviewed-by: Dan Mick <dan.mick@inktank.com>
| * | | | | | | | | | doc: Added sudo to the service start command.John Wilkins2012-12-071-1/+1
| |/ / / / / / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
| * | | | | | | | | doc: Moved sudo to before ssh instead of before tee.John Wilkins2012-12-071-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: John Wilkins <john.wilkins@inktank.com>
View Lorry Controller Status