summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* crush: use breadth-first search for indep modewip-crushSage Weil2013-08-112-10/+156
| | | | Signed-off-by: Sage Weil <sage@inktank.com>
* crush: return CRUSH_ITEM_UNDEF for failed placements with indepSage Weil2013-08-112-2/+8
| | | | | | | | | For firstn mode, if we fail to make a valid placement choice, we just continue and return a short result to the caller. For indep mode, however, we need to make the position stable, and return an undefined value on failed placements to avoid shifting later results to the left. Signed-off-by: Sage Weil <sage@inktank.com>
* crush: eliminate CRUSH_MAX_SET result size limitationSage Weil2013-08-114-7/+22
| | | | | | | | This is only present to size the temporary scratch arrays that we put on the stack. Let the caller allocate them as they wish and remove the limitation. Signed-off-by: Sage Weil <sage@inktank.com>
* crush: fix commentSage Weil2013-08-111-1/+1
| | | | Signed-off-by: Sage Weil <sage@inktank.com>
* client: add enclosing object for asok dumpsSage Weil2013-08-111-0/+2
| | | | Signed-off-by: Sage Weil <sage@inktank.com>
* doc/release-notes: adjust whitespaceSage Weil2013-08-091-45/+45
| | | | Signed-off-by: Sage Weil <sage@inktank.com>
* doc/release-notes: v0.67 release notesSage Weil2013-08-092-45/+160
| | | | Signed-off-by: Sage Weil <sage@inktank.com>
* Merge remote-tracking branch 'gh/next'Sage Weil2013-08-099-30/+64
|\
| * ceph: retry new-style mon command if we get EINVAL from the old-style commandSage Weil2013-08-091-15/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We can race with a mon upgrade: - get command descriptions, get EINVAL - mons upgrade, new quorum - send old-style command - get EINVAL In this case, we should try one last time to get the command descriptions. Fixes: #5788 Reviewed-by: Dan Mick <dan.mick@inktank.com> Signed-off-by: Sage Weil <sage@inktank.com>
| * common: pick_addresses: fix bug with observer class that triggered #5205Joao Eduardo Luis2013-08-091-7/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The Observer class we defined to observe conf changes and thus avoid triggering #5205 (as fixed by eb86eebe1ba42f04b46f7c3e3419b83eb6fe7f9a), was returning always the same const static array, which would lead us to always populate the observer's list with an observer for 'public_addr'. This would of course become a problem when trying to obtain the observer for 'cluster_add' during md_config_t::set_val() -- thus triggering the same assert as initially reported on #5205. Backport: cuttlefish Fixes: #5205 Signed-off-by: Joao Eduardo Luis <jecluis@gmail.com> Reviewed-by: Sage Weil <sage@inktank.com>
| * rgw: return 423 Locked response when failing to lock objectYehuda Sadeh2013-08-094-0/+8
| | | | | | | | | | | | | | | | | | | | Fixes: #5882 Translate the EBUSY we get when trying to lock a shard / object to 423 Locked response. Beforehand it was just translated to the default 500. Reviewed-by: Sage Weil <sage@inktank.com> Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * rgw: make RGWHTTPClient callback methods pure virtualYehuda Sadeh2013-08-092-3/+22
| | | | | | | | | | | | | | | | This will make it easier to catch interface changes, like the ones that triggered #5921. Reviewed-by: Sage Weil <sage@inktank.com> Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * rgw: rename read_header() to receive_header() where neededYehuda Sadeh2013-08-091-2/+2
| | | | | | | | | | | | | | The same as with #5921, fixing a missed interface change. Reviewed-by: Sage Weil <sage@inktank.com> Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * rgw: rename data receive callbacks in swift token revocationYehuda Sadeh2013-08-091-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | Fixes: #5921 As part of the work that was made for dumpling, the http client in-data callback was renamed in order to avoid confusion. However, we missed the rename in a couple of places, which this patch amend. Reported-by: Roald van Loon <roaldvanloon@gmail.com> Reviewed-by: Sage Weil <sage@inktank.com> Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * RadosClient: shutdown monclient after dropping lockSamuel Just2013-08-081-1/+1
| | | | | | | | | | | | | | | | | | Otherwise, the monclient shutdown may deadlock waiting on a context trying to take the RadosClient lock. Fixes: #5897 Signed-off-by: Samuel Just <sam.just@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
* | Merge pull request #486 from kri5/wip-s3-compliance-docYehuda Sadeh2013-08-092-1/+306
|\ \ | | | | | | | | | | | | Wip s3 compliance doc Reviewed-by: Yehuda Sadeh <yehuda@inktank.com>
| * | doc: Add a page about rgw S3 API complianceChristophe Courtaut2013-08-092-1/+306
| | | | | | | | | | | | Signed-off-by: Christophe Courtaut <christophe.courtaut@gmail.com>
* | | Merge remote-tracking branch 'gh/wip-5648-c'Sage Weil2013-08-0910-145/+533
|\ \ \ | | | | | | | | | | | | Reviewed-by: Sage Weil <sage@inktank.com>
| * | | test: mon: moncap: add 'allow command foo with arg="bar.baz"' testsJoao Eduardo Luis2013-08-061-0/+2
| | | | | | | | | | | | | | | | Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
| * | | qa: workunits: mon: workunit to stress the monitor's capsJoao Eduardo Luis2013-08-061-0/+370
| | | | | | | | | | | | | | | | Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
| * | | mon: AuthMonitor: change 'auth add' behaviourJoao Eduardo Luis2013-08-061-16/+98
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If an entity already existed, 'auth add' would smash its key and caps with whatever was on the supplied keyring file; if a keyring weren't specified, we would simply generate a new key and destroy all existing caps (unless caps were specified and happened to be different from the already in-place caps). This behaviour is obviously sketchy. With this patch we now enforce the following behaviour: - if entity does not exist in current state, check if we are about to create it (by checking the pending state); if so, wait for the new state to be committed and re-handle the command then, so we don't get bad results from pending request - if the command reproduces the current state (same key, same caps), we return 0; else, - if entity exists and supplied key OR caps are different, return -EINVAL - else create a new entity. Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
| * | | mon: drop old _allowed_command()Sage Weil2013-08-062-42/+0
| | | | | | | | | | | | | | | | Signed-off-by: Sage Weil <sage@inktank.com>
| * | | mon: services: no longer needed to enforce caps on a per-service basisJoao Eduardo Luis2013-08-067-31/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | We now perform all perm checks for commands on Monitor::handle_command(). Services no longer need to check them. Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
| * | | mon: Monitor: purge old permission checking format on handle_command()Joao Eduardo Luis2013-08-061-57/+0
| | | | | | | | | | | | | | | | Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
| * | | mon: Monitor: check caps considering command's requirementsJoao Eduardo Luis2013-08-062-0/+52
| | | | | | | | | | | | | | | | | | | | | | | | Fixes: #5648 Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
* | | | PendingReleaseNotes: note 'ceph daemon ...' argument behavior changeSage Weil2013-08-091-0/+11
| |/ / |/| | | | | | | | Signed-off-by: Sage Weil <sage@inktank.com>
* | | mds: notify clients about deleted inodeYan, Zheng2013-08-081-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | To make sure clients trim the deleted inode from the their cache ASAP. After all clients release the inode, we can reclaim space. Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com> Reviewed-by: Sage Weil <sage@inktank.com>
* | | Merge remote-tracking branch 'yan/wip-mds'Sage Weil2013-08-087-56/+52
|\ \ \ | | | | | | | | | | | | Reviewed-by: Sage Weil <sage@inktank.com>
| * | | mds: don't issue caps while session is staleYan, Zheng2013-08-051-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Locker::issue_caps() does not issue new caps to stale client, CInode::encode_inodestat() should have the same logical. Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
| * | | mds: change LOCK_SCAN to unstable stateYan, Zheng2013-08-054-16/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | commit 0071b8e75b (mds: stay in SCAN state in file_eval) makes Locker::file_eval() ignore lock in LOCK_SCAN state. If there is no request changes the lock state, the lock can be stuck in LOCK_SCAN state forever. This can cause client read/write hang because lock in LOCK_SCAN state does not allow Frw caps. The fix is change LOCK_SCAN to a unstable state. Thank to the CInode::STATE_RECOVERING check in Locker::eval_gather(), the lock stays in the SCAN state while file is being recovering. The lock will transit to a stable state once the recovery finishes. Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
| * | | mds: handle "state == LOCK_LOCK_XLOCK" when cancelling xlockYan, Zheng2013-08-051-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If we find lock state is LOCK_LOCK_XLOCK when cancelling xlock, set lock state to LOCK_XLOCK_DONE and call Locker::eval_gather(). This makes sure the lock will eventually transit to a stable state. (LOCK_XLOCK_DONE's next state is stable) Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
| * | | mds: remove "type != CEPH_LOCK_DN" check in Locker::cancel_locking()Yan, Zheng2013-08-051-7/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For acquiring/cancelling xlock, the lock state transitions for dentry lock and other types of locks are the same. So I think the "type != CEPH_LOCK_DN" check doesn't make sense. Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
| * | | mds: revoke GSHARED cap when finishing xlockYan, Zheng2013-08-052-32/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If lock state is LOCK_XLOCKDONE, the xlocker can have GSHARED cap. So when finishing xlock, we may need to revoke the GSHARED cap. In most cases Locker::_finish_xlock() directly set lock state to LOCK_LOCK or LOCK_EXCL, which hides the issue. If 'num_rdlock > 0' or 'num_wrlock > 0' when finishing xlock, the issue reveals. (lock get stuck in LOCK_XLOCKDONE forever) The fix is always call Locker::_finish_xlock() when xlock count reaches zero. _finish_xlock() checks if it can change lock state to LOCK_EXCL immediately. If not, it uses Locker::eval_gather() to transit lock state. Another change of this patch is avoid changing lock state to LOCK_LOCK directly. because lock in LOCK_XLOCK_DONE state allows GSHARED cap, lock in LOCK_LOCK state does not. Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
| * | | mds: fix cap revoke confirmationYan, Zheng2013-08-051-3/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are several issues in the Capability::confirm_receipt() 1. when receiving a client caps message with 'seq == last_sent', it doesn't mean we finish revoking caps. The client can send caps message that only flushes dirty metadata. 2. When receiving a client caps message with 'seq == N', we should forget pending revocations whose seq numbers are less than N. This is because, when revoking caps, we create a revoke_info structure and set its seq number to 'last_sent', then increase the 'last_sent'. 3. When client actively releases caps (by request), the code only works for the 'seq == last_sent' case. If there are pending revocations, we should update them as if the release message is received before we revoke the corresponding caps. Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
* | | | Merge remote-tracking branch 'gh/next'Sage Weil2013-08-0835-167/+386
|\ \ \ \ | | |_|/ | |/| |
| * | | buffer: change #include orderSage Weil2013-08-081-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | huangjun <hjwsm1989@gmail.com> reports that this fixes the compilation on CentOS5. Signed-off-by: Sage Weil <sage@inktank.com>
| * | | make sure we are using the mount optionsAlfredo Deza2013-08-081-0/+4
| | | | | | | | | | | | | | | | | | | | Signed-off-by: Alfredo Deza <alfredo@deza.pe> (cherry picked from commit 34831d0989d4bcec4920068b6ee09ab6b3234c91)
| * | | Merge pull request #487 from ceph/wip-5746Sage Weil2013-08-081-3/+8
| |\ \ \ | | | | | | | | | | | | | | | | | | | | ceph.in: Re-enable ceph interactive mode (missing its output). Reviewed-by: Sage Weil <sage@inktank.com>
| | * | | ceph.in: Re-enable ceph interactive mode (missing its output).Dan Mick2013-08-081-3/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Also, loop on error. There's no reason to exit the interpreter loop on an error, and it's probably less annoying if we don't. Print the error, and any output, and continue. Fixes: #5746 Signed-off-by: Dan Mick <dan.mick@inktank.com>
| * | | | Merge pull request #488 from dmick/wip-dmick-daemon-einvalSage Weil2013-08-081-2/+4
| |\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | Wip dmick daemon einval Reviewed-by: Sage Weil <sage@inktank.com>
| | * | | | ceph.in: return EINVAL on daemon command errorDan Mick2013-08-081-2/+4
| |/ / / / | | | | | | | | | | | | | | | Signed-off-by: Dan Mick <dan.mick@inktank.com>
| * | | | mon: fix 'osd crush rule rm ...' dup argSage Weil2013-08-081-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This was broken way back in 0d66c9ebbf626117c641c975a8682a0aaba588c4, but we were ignoring the dup until recently. t Signed-off-by: Sage Weil <sage@inktank.com>
| * | | | qa/workunits/cephtool/test.sh: test set/unset of all osd flagsSage Weil2013-08-071-0/+8
| | | | | | | | | | | | | | | | | | | | Signed-off-by: Sage Weil <sage@inktank.com>
| * | | | mon/MonCommands: fix typo nobackfile -> nobackfillSage Weil2013-08-071-2/+2
| | | | | | | | | | | | | | | | | | | | Signed-off-by: Sage Weil <sage@inktank.com>
| * | | | Add back the mistakenly removed "ceph osd set noscrub/nodeep-scrub"Dan Mick2013-08-072-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fixes: #5868 Signed-off-by: Dan Mick <dan.mick@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
| * | | | Merge branch 'wip-recovery-op-warn' into nextSamuel Just2013-08-069-50/+100
| |\ \ \ \ | | | | | | | | | | | | | | | | | | Reviewed-by: Sage Weil <sage@inktank.com>
| | * | | | OSD: suspend timeout on process_peering_events pg lockSamuel Just2013-08-061-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
| | * | | | OSD: only start osd_recovery_max_single_start at onceSamuel Just2013-08-062-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
| | * | | | ReplicatedPG: ping tphandle between pushesSamuel Just2013-08-062-7/+11
| | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
| | * | | | ReplicatedPG: ping TPHandle during scan_rangeSamuel Just2013-08-066-22/+56
| | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
View Lorry Controller Status