summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* ReplicatedPG,PG: use PGBackend methods for getattrwip-5992Samuel Just2013-09-192-16/+28
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG,PG: adapt collection_list* users to PGBackend interfaceSamuel Just2013-09-192-20/+20
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* PG,ReplicatedPG: expose PGBackend to PGSamuel Just2013-09-192-0/+6
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* PG.cc: remove leading empty spaceSamuel Just2013-09-191-1/+0
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* PGBackend,ReplicatedBackend: add interfaces for scanning the pgSamuel Just2013-09-193-0/+89
| | | | | | | This will be important since the erasure coded pg will have a different on-disk format than the replicated backend. Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: don't rescan the local collection if we can avoid itSamuel Just2013-09-193-8/+49
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG.h: remove temp_coll stubSamuel Just2013-09-191-5/+0
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* OSD,ReplicatedPG: queue pushes in a op_tp workqueueSamuel Just2013-09-193-4/+11
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: don't proceed to backfill until recovering is emptySamuel Just2013-09-191-2/+5
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: correctly handle backfillSamuel Just2013-09-193-18/+18
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: fix up recalc_subsetsSamuel Just2013-09-193-35/+16
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedBackend: wire in start_pushesSamuel Just2013-09-193-10/+71
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedBackend: handle stats properlySamuel Just2013-09-194-24/+38
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedBackend: track object contexts in Push/PullInfoSamuel Just2013-09-192-15/+7
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedBackend: remove priority from Pull/PushInfoSamuel Just2013-09-192-3/+1
| | | | | | Instead, we'll just nab the priority from the messages. Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: move prep_push and friends to ReplicatedBackendSamuel Just2013-09-194-110/+115
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: remove pushing/pullingSamuel Just2013-09-192-73/+19
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: handle down pull sourcesSamuel Just2013-09-195-18/+39
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: split recover_missing out of prepare_pullSamuel Just2013-09-194-108/+180
| | | | | | Also, move prepare_pull to PGBackend. Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: shuffle over submit_push_*/build_push_opSamuel Just2013-09-193-20/+20
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: move over send_pushes/send_pulls/prep_push_op_blankSamuel Just2013-09-193-32/+27
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: move over _failed_pushSamuel Just2013-09-193-4/+10
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: move handle_*Samuel Just2013-09-193-32/+25
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: shuffle do_pull/do_push_replySamuel Just2013-09-193-6/+4
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: move _do_push,_do_pull_response,do_pushSamuel Just2013-09-193-21/+17
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: convert sub_op_pullSamuel Just2013-09-193-3/+2
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: make log_subop_stats staticSamuel Just2013-09-192-25/+22
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: shuffle sub_op_push_replySamuel Just2013-09-193-3/+2
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG/Backend: shuffle sub_op_push over to ReplicatedBackendSamuel Just2013-09-194-17/+19
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* osd/: allow dout for ReplicatedBackend methods to work in ReplicatedPG.ccSamuel Just2013-09-191-2/+3
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: extract PGBackend::Listener recovery callbacksSamuel Just2013-09-194-109/+159
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* PG,ReplicatedPG: handle do_request in ReplicatedPG,PGBackendSamuel Just2013-09-195-89/+68
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: pass on dump_recovery_info to PGBackendSamuel Just2013-09-191-43/+6
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* OSD,ReplicatedPG: let PGBackend handle the temp collectionSamuel Just2013-09-197-77/+69
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* osd/: add PGBackend interfaces and stubsSamuel Just2013-09-198-2/+726
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: Allow get_object_context caller to provide attributesSamuel Just2013-09-193-32/+65
| | | | | | | This will be used by PGBackend implementers to get the SnapSet and ObjectInfo for newly recovered objects on the primary. Signed-off-by: Samuel Just <sam.just@inktank.com>
* ReplicatedPG: send_push is unusedSamuel Just2013-09-192-16/+0
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* FileStore: ping tphandle between each operation within a transactionSamuel Just2013-09-192-3/+13
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* WorkQueue: add a workqueue which simply runs queued GenContextsSamuel Just2013-09-191-0/+43
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* Context: add GenContext templated on the callback argumentSamuel Just2013-09-191-0/+20
| | | | Signed-off-by: Samuel Just <sam.just@inktank.com>
* Makefile: fix unittest_crc32cSage Weil2013-09-191-0/+5
| | | | | | This got dropped during the makefile refactor. Signed-off-by: Sage Weil <sage@inktank.com>
* os/FileStore: fix uninitialized varSage Weil2013-09-181-1/+1
| | | | Signed-off-by: Sage Weil <sage@inktank.com>
* doc/release-notes: v0.69Sage Weil2013-09-181-0/+50
| | | | Signed-off-by: Sage Weil <sage@inktank.com>
* Merge branch 'next'Gary Lowell2013-09-1811-6/+36
|\
| * v0.69v0.69Gary Lowell2013-09-182-1/+7
| |
| * rgw: try to create log pool if doesn't existYehuda Sadeh2013-09-131-0/+9
| | | | | | | | | | | | | | | | When using replica log, if the log pool doesn't exist all operations are going to fail. Try to create it if doesn't exist. Reviewed-by: Josh Durgin <josh.durgin@inktank.com> Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * rgw: NULL terminate buffer before parsing itYehuda Sadeh2013-09-131-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | Fixes: #6175 Backport: dumpling We get a buffer off the remote gateway which might not be NULL terminated. The JSON parser needs the buffer to be NULL terminated even though we provide a buffer length as it calls strlen(). Reviewed-by: Josh Durgin <josh.durgin@inktank.com> Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * rgw: don't call list::size() in ObjectCacheYehuda Sadeh2013-09-112-2/+6
| | | | | | | | | | | | | | | | Fixes: #6286 Use an external counter instead of calling list::size() Reviewed-by: Sage Weil <sage@inktank.com> Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * rgw: drain pending requests before completing writeYehuda Sadeh2013-09-112-1/+6
| | | | | | | | | | | | | | | | | | | | Fixes: #6268 When doing aio write of objects (either regular or multipart parts) we need to drain pending aio requests. Otherwise if gateway goes down then object might end up corrupted. Reviewed-by: Josh Durgin <josh.durgin@inktank.com> Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
| * rgw: when failing read from client, return correct errorYehuda Sadeh2013-09-101-1/+1
| | | | | | | | | | | | | | | | | | | | | | Fixes: #6214 When getting a failed read from client when putting an object we returned the wrong value (always 0), which in the chunked- upload case ended up in assuming that the write was done successfully. Signed-off-by: Yehuda Sadeh <yehuda@inktank.com> Reviewed-by: Sage Weil <sage@inktank.com>
View Lorry Controller Status