summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* 16.2.0-90-g50f1821b4cHEADmasterKefu Chai2021-04-123230-0/+92
| | | | Signed-off-by: Kefu Chai <kchai@redhat.com>
* mark some in-cluster messageKefu Chai2021-02-248-0/+0
| | | | | | | | | | | | | | | | | | | | mark following messages forward_incompat * MMgrBeacon * MOSDBoot * MOSDFailure * MOSDPGLog * MOSDRepOp * MOSDRepOpReply * MForward (as the container of the messages above) * LogEntry as there is not need to decode these in-cluster messages if they are encoded by pre-nautilus daemons. in quincy, we only need to decode in-cluster messages generated by octopus and pacific daemons. so it's safe to stop decoding these 11.0 (kraken) messages. Signed-off-by: Kefu Chai <tchaikov@gmail.com>
* remove archive/17.0.0-200/MOSDScrubKefu Chai2021-02-241-0/+0
| | | | | | | | | | | | | | | | since we don't have any corpus for MOSDScrub, there is no need to create this file. also we should create the file named $type under "forward_incompat". so, if we have had some corpus for MOSDScrub, we could create a file at archive/15.2.1-40-ga838bb1aae/objects/MOSDScrub assuming the corpus of MOSDScrub is located under archive/15.2.1-40-ga838bb1aae/objects. this change is a follow-up of 20e09990f9faf8998da8d2ee403393043be6405d Signed-off-by: Kefu Chai <tchaikov@gmail.com>
* ceph-object-corpus: forward incompat log related classesJoao Eduardo Luis2021-02-193-0/+0
| | | | Signed-off-by: Joao Eduardo Luis <joao@suse.com>
* mark MOSDScrub forward_incompat as of v17.0.0-200Sage Weil2021-01-281-0/+0
| | | | Signed-off-by: Sage Weil <sage@newdream.net>
* slave -> peerSage Weil2020-07-149-0/+0
| | | | Signed-off-by: Sage Weil <sage@newdream.net>
* archive: mark EMetaBlob denc incompatible release before 60730271d4Kefu Chai2020-07-1127-0/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | in https://github.com/ceph/ceph/commit/60730271d49a3c16bee6ff60c402b924a0280efc, the encoding schema version of EMetaBlob bumped from 4 to 5. the older release containing this change was v0.58 in https://github.com/ceph/ceph/commit/d62c67ca2041797fc7aa212ae4a37ba91e5620e7, the struct_v of EMetaBlob::fullbit::encode was bumped from 7 to 8, , the oldest release containing this change was hammer rc1, i.e., v0.93. in https://github.com/ceph/ceph/pull/35931/commits/0127f2b01928b515bbbb7811976012ec97d3a957, the backward compatible code of EMetaBlob::fullbit was dropped, so it does not decode any EMetaBlob::fullbit whose struct_v belongs to [2, 8) anymore. in this change, all EMetaBlob* corpus created before 9.3 or 0.93 are marked forward incompatible, so the denc tests with them are skipped if the dencoder version is higher than those versions. this change helps us to drop the backward compatible code to decode EMetaBlob. as it is not neeed anymore after 2 releases. Signed-off-by: Kefu Chai <tchaikov@gmail.com>
* Added octopus 15.2.1-40-ga838bb1aaeJosh Durgin2020-04-163042-0/+90
| | | | Signed-off-by: Josh Durgin <jdurgin@redhat.com>
* Merge pull request #8 from letterwuyu/forward_incompat_pg_missingJosh Durgin2019-05-082-0/+0
|\ | | | | | | | | ceph-object-corpus: forward_incompat pg_missing_item and pg_missing_t Reviewed-by: Josh Durgin <jdurgin@redhat.com>
| * ceph-object-corpus: forward_incompat pg_missing_item and pg_missing_troot2019-05-092-0/+0
|/ | | | | | | since luminous, recovery_deletes is always true, and octopus won't maintain compatibility with jewel OSDs Signed-off-by: lishuhao <lishuhao@unitedstack.com>
* Merge pull request #6 from tchaikov/wip-15.0.0-539Sage Weil2019-04-242761-0/+93
|\ | | | | 15.0.0-539-g191ab33faf
| * 15.0.0-539-g191ab33fafKefu Chai2019-04-242761-0/+93
|/
* Merge pull request #5 from batrick/MMDSResolveSage Weil2018-04-041-0/+0
|\ | | | | archive/*/objects/MMDSResolve: remove MMDSResolve messages
| * archive/13.0.0/forward_incompat: add MMDSResolvePatrick Donnelly2018-04-041-0/+0
|/ | | | | | | | | These messages are versioned via CEPH_MDS_PROTOCOL cluster protocol. The decoder doesn't know how to handle older messages yet. Relevant ceph change: ebd0fcd067 Signed-off-by: Patrick Donnelly <batrick@batbytes.com>
* archive/12.0.0/forward_incompat: add PGMapSage Weil2017-08-301-0/+0
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* mark pre-luminous messages incompatwip-pre-luminousSage Weil2017-08-245-0/+0
|
* Merge pull request #4 from tchaikov/wip-MMDSSlaveRequestSage Weil2017-06-162-0/+0
|\ | | | | add MMDSSlaveRequest incompat at 0.80-rc1-35-g4812150
| * add MMDSSlaveRequest incompat at 0.80-rc1-35-g4812150Kefu Chai2017-06-162-0/+0
|/ | | | | | we added "mds_rank_t srcdn_auth" to it in a1842eb2 Signed-off-by: Kefu Chai <tchaikov@gmail.com>
* Merge remote-tracking branch 'origin/kraken'Sage Weil2017-05-053402-0/+193
|\
| * kraken 11.2.0-280-g34e04dekrakenSage Weil2017-05-053402-0/+193
| |
* | remove decoding compat for misc pre-10.2.0 formatsSage Weil2017-04-2423-0/+0
| | | | | | | | | | | | | | All upgrades pass through jewel, so we can remove compat for various on-wire types and on-disk items that we know will be rewritten. Signed-off-by: Sage Weil <sage@redhat.com>
* | add ScrubMap[::object] incompat at 10.2.0Sage Weil2017-02-272-0/+0
|/ | | | | | | | We're removing decoding ability for older instances because we no longer need to read anything pre-jewel (this is an on-wire message and upgrades pass through jewel). Signed-off-by: Sage Weil <sage@redhat.com>
* drop 0.46-add-osd-uuid/objects/OSDMap/9wip-readable-fixesSage Weil2017-02-271-0/+0
| | | | | | | | | This is the one OSDMap instance with stray trailing data, which as of 3f3a72b42e6410c13dc98d6b05f53ec9da93c4fe ceph-dencoder no longers likes for OSDMap. Easier to remove this ancient test instance than add the new macros needed for STRAYDATA + NOCOPY. Signed-off-by: Sage Weil <sage@redhat.com>
* ECSubWrite embeds ObjectStore::TransactionSage Weil2017-02-271-0/+0
| | | | | | So it's incompat as of the same version. Signed-off-by: Sage Weil <sage@redhat.com>
* Merge pull request #2 from aiicore/drop-removal-pg-typeSage Weil2015-12-162-0/+2
|\ | | | | Drop deprecated removal pg type.
| * Drop deprecated removal pg type.Igor Podoski2015-12-162-0/+2
|/ | | | | | | | Two testing objects from coll_t type need to be removed. Those objects consists FORREMOVAL_ pg type which is deprecaded. Signed-off-by: Igor Podoski <igor.podoski@ts.fujitsu.com>
* 0.94.2: remove coll_t's foo and barSage Weil2015-08-272-0/+0
| | | | | | | These were only generated from tests, not clusters, and they fail to decode in infernalis. Signed-off-by: Sage Weil <sage@redhat.com>
* 0.94.2-207-g88e7ee7wip-osd-compatSage Weil2015-08-271783-0/+4059
|
* new versions can no longer decode some pre-hammer ObjectStore::TransactionSage Weil2015-08-071-0/+0
| | | | Specifically the collection attr operations are no longer supported.
* drop coll_t 'foo' and 'bar'coll-foo-barSage Weil2015-01-104-0/+0
| | | | | | | | These are no longer valid, but we want to keep testing all of the other old encoded instances. Just drop them since we don't care about whether they decode in real clusters. Signed-off-by: Sage Weil <sage@redhat.com>
* drop compat with very old ObjectStore::Transactionwip-fast-txnSage Weil2014-12-111-0/+0
| | | | Signed-off-by: Sage Weil <sage@redhat.com>
* MDiscoverReply encoding changedfooSage Weil2014-05-031-0/+0
|
* 0.80-rc1-35-g4812150fireflySage Weil2014-05-031334-0/+56
| | | | Signed-off-by: Sage Weil <sage@inktank.com>
* remove 0.41*, 0.43*Sage Weil2014-02-211737-72/+0
| | | | | | These are old releases and we cover these object types later. Signed-off-by: Sage Weil <sage@inktank.com>
* note MonCap incompat around 0.63Sage Weil2013-07-121-0/+0
|
* Merge remote-tracking branch 'cc/cuttlefish'Sage Weil2013-07-121536-0/+53
|\
| * 0.61.4-60-g24c59becuttlefishSage Weil2013-07-121536-0/+53
| |
* | note inode_backpointer_t incompat back in 0.56-629Sage Weil2013-07-111-0/+0
| | | | | | | | Signed-off-by: Sage Weil <sage@inktank.com>
* | 0.66-500-g93b3e63Sage Weil2013-07-111525-0/+52
| |
* | note MMonSync incompat change with new scrubSage Weil2013-07-111-0/+0
| |
* | MForward incompatwip-mon-capSage Weil2013-05-091-0/+0
|/ | | | | | | We changed MonCap[s] encoding and bumped the protocol. Note the incompat chagne. Signed-off-by: Sage Weil <sage@inktank.com>
* note forward_incompat changes for single-paxos at 0.54-2289-gbeafca5wsp.master.newJoao Eduardo Luis2013-02-212-0/+0
| | | | Signed-off-by: Joao Eduardo Luis <joao.luis@inktank.com>
* v0.52 objectsSage Weil2012-09-20989-0/+56
| | | | (Actually, these were generated just before v0.52 release.)
* add MonMap v2Sage Weil2012-05-211-0/+0
|
* OSDMap and OSDMap::Incremental instances with new osd uuid fields.Sage Weil2012-05-0218-0/+0
|
* add prune-archive.shSage Weil2012-04-181-0/+5
|
* prune archiveSage Weil2012-04-183230-322/+0
| | | | 10 objects per type, version
* adding corpus for 0.43Yehuda Sadeh2012-03-093135-0/+262
| | | | Signed-off-by: Yehuda Sadeh <yehuda.sadeh@dreamhost.com>
* note forward_incompat changes from osd_peer_stat_t at 0.41-306-gfbbbd01Sage Weil2012-02-133-0/+0
| | | | | | | | | osd_peer_stat_t isn't in the old corpus, but it was contained by messages that used it; mark those as forward_incompat (meaning new code won't understand the old data). We did this because it's an on-wire incompatibility only, and a new feature bit protects us there.
* restructure hierarchySage Weil2012-02-131829-0/+0
| | | | archive/$version/objects/$type/...
View Lorry Controller Status