Evan Tschannen
|
126fc53d10
|
fix: the start version for peek cursors that merge with multiple log sets is the maximum of the individual start versions
|
2018-04-23 12:42:51 -07:00 |
|
Evan Tschannen
|
73597f190e
|
fix: new tlogs are initialized with exactly the tags which existed at the recovery version
|
2018-04-22 20:28:01 -07:00 |
|
Evan Tschannen
|
a520d03397
|
fix: if we cannot find a tag, it must have been popped at the recovery version.
|
2018-04-22 15:08:38 -07:00 |
|
Evan Tschannen
|
ef23136809
|
fix: ensure the logSystemConfig is updated with newly recruited log routers
|
2018-04-22 11:54:39 -07:00 |
|
Evan Tschannen
|
fceec020de
|
fix: use the known committed version if the last generation primary logs were in the same data center as this generation
the known committed version in end epoch is the maximum seen in all responses regardless of log set
|
2018-04-22 11:14:13 -07:00 |
|
Evan Tschannen
|
c3a344d44e
|
fix: do not choose a remote start version past the start of the locked logs
|
2018-04-21 16:03:28 -07:00 |
|
Evan Tschannen
|
28a1fa9dc2
|
fix: we need to notify the old log system that its recruitmentID has changed
|
2018-04-21 12:57:00 -07:00 |
|
Evan Tschannen
|
1d1e2cd367
|
fix: initialize the known committed version on the tlog
|
2018-04-21 00:41:15 -07:00 |
|
Evan Tschannen
|
8d350ceb5f
|
fix: persist the known committed version on the tlogs
|
2018-04-20 17:55:46 -07:00 |
|
Evan Tschannen
|
a6d9e889f0
|
a cleaner solution to preventing tlogs from peeking log routers
|
2018-04-20 13:25:22 -07:00 |
|
Evan Tschannen
|
f5c3417905
|
fix: prevent tlogs from peeking the wrong log routers
|
2018-04-20 00:30:37 -07:00 |
|
Evan Tschannen
|
5da452db8e
|
fix: pop the log routers again after the log system updates
|
2018-04-19 14:33:31 -07:00 |
|
Evan Tschannen
|
22526ef996
|
fix: do not tell storage servers about large sections of empty versions, because it can lead them to make mutations durable which have not been committed
|
2018-04-18 16:06:44 -07:00 |
|
Evan Tschannen
|
447c7bd15b
|
fix: log routers use durable known committed version at the time of the pop to determine what is safe to pop from their logs
fix: storage server does not advance its version across large version increase until it has data associated with the version
|
2018-04-18 12:07:29 -07:00 |
|
Evan Tschannen
|
e43fb6d8bc
|
fix: the log routers were popping too many versions because the known committed version is less than minPopped version
|
2018-04-17 19:41:36 -07:00 |
|
Evan Tschannen
|
8569a85771
|
fix: only let a log router pop if they tlog it is serving is fully recovered
|
2018-04-17 15:03:22 -07:00 |
|
Evan Tschannen
|
760bc8bc99
|
fix: log router version needs to be fetched before it is available
fix: tlog did not fetch known committed version if start version was exactly equal to it
|
2018-04-17 11:16:48 -07:00 |
|
Evan Tschannen
|
093908b83f
|
fix: log routers were starting one version too late
|
2018-04-17 00:29:16 -07:00 |
|
Evan Tschannen
|
3e40505f4a
|
Revert "fix: remote logs should reply until they have recovered through recoverAt"
This reverts commit 3c0c03c004.
|
2018-04-16 23:17:16 -07:00 |
|
Evan Tschannen
|
3c0c03c004
|
fix: remote logs should reply until they have recovered through recoverAt
|
2018-04-16 17:25:49 -07:00 |
|
Evan Tschannen
|
cef6c9b418
|
fix: the startVersion cannot be larger than the known committed version
|
2018-04-16 16:21:27 -07:00 |
|
Evan Tschannen
|
dcfa1847ff
|
fix: log router’s starting popped version must be less than its starting version
|
2018-04-16 11:43:03 -07:00 |
|
Evan Tschannen
|
3018a7b1b3
|
fix: the known committed version of a newly initialized log is 1, since by definition the first commit must have succeeded
|
2018-04-16 10:42:48 -07:00 |
|
Evan Tschannen
|
a8662f8737
|
fix: remote recovered is does not need to wait for old logs to be removed
|
2018-04-16 10:14:39 -07:00 |
|
Evan Tschannen
|
e53f17a83a
|
fix: the newest log router needs to start where the last old one ends
|
2018-04-15 14:54:22 -07:00 |
|
Evan Tschannen
|
5533016f1e
|
fix: tlogs are now initialized immediately, instead of when starting the core, this must be done to pop the log routers during recovery
fix: log router start version must be the same as remote log start version
|
2018-04-15 14:33:07 -07:00 |
|
Evan Tschannen
|
0496bee1ef
|
fix: suppress expected errors in data distribution
|
2018-04-15 11:30:22 -07:00 |
|
Evan Tschannen
|
041f5787fb
|
fix: peekLocal does not stop when a locality does not exist
fix: lock logs only stops on special or upgraded locality
fix: recruiting old log routers respects the passed in startVersion
|
2018-04-14 19:06:24 -07:00 |
|
Evan Tschannen
|
f5141acae9
|
fix: log routers need all logs present in their log system since they call addRemoteTags
|
2018-04-13 17:33:36 -07:00 |
|
Evan Tschannen
|
65e69620a7
|
fix: unrecoveredBefore on a new log is at minimum 1
|
2018-04-13 10:41:30 -07:00 |
|
Evan Tschannen
|
c589630e53
|
fix: log router start version is based on the start version of the local logs
|
2018-04-12 18:14:23 -07:00 |
|
Evan Tschannen
|
3b7e4410cf
|
fix: protect from peeking too early of a version from a log router
|
2018-04-12 16:15:17 -07:00 |
|
Evan Tschannen
|
1af5ac0d9d
|
fix: a number of different problems prevented tlogs from using log routers during recovery
|
2018-04-12 15:20:54 -07:00 |
|
Evan Tschannen
|
c6229e443c
|
fix: do not use resolution class when using regions
|
2018-04-11 21:22:53 -07:00 |
|
Evan Tschannen
|
4248fbec61
|
fix: must set startVersion when upgrading
|
2018-04-11 17:33:17 -07:00 |
|
Evan Tschannen
|
3453a51d0f
|
remoteRecovery was still swallowing errors
|
2018-04-10 13:31:24 -07:00 |
|
Evan Tschannen
|
5fcedd2e98
|
fix: coordinated state errors were being eaten
|
2018-04-10 11:14:57 -07:00 |
|
Evan Tschannen
|
2ab2c788b3
|
fix: the start version is allowed to be larger than the recovery version
|
2018-04-09 21:58:14 -07:00 |
|
Evan Tschannen
|
a738c4bec1
|
fix: if the known committed version is equal to the recovery version we do not need to copy any data
|
2018-04-09 20:48:55 -07:00 |
|
Evan Tschannen
|
419951f601
|
fix: need to initialize tlog versions to less than the startVersion
|
2018-04-09 17:17:11 -07:00 |
|
Evan Tschannen
|
27e14790b1
|
fix: do not start at a version larger that the recovery version
|
2018-04-09 15:08:01 -07:00 |
|
Evan Tschannen
|
7566a0d109
|
fix: endEpoch gets its logs from the core state, so by definition they are written
|
2018-04-09 11:44:54 -07:00 |
|
Evan Tschannen
|
4c89f721cd
|
fix: do not include logRouter tags in lock results
|
2018-04-09 10:48:57 -07:00 |
|
Evan Tschannen
|
7af892f50b
|
first working version of non-copying recovery working with fearless configurations
|
2018-04-08 21:24:05 -07:00 |
|
Evan Tschannen
|
331e707684
|
fix: pop all tags that did not have data at the recovery version because fully popped tags may come back when pullAsyncData re-indexes the mutations
|
2018-03-31 16:47:56 -07:00 |
|
Evan Tschannen
|
96fffe2cea
|
fix: do not update version if the log has been stopped
|
2018-03-30 22:11:42 -07:00 |
|
Evan Tschannen
|
4fb2b99341
|
fix: using only one region still means we need 3 machines per datacenter, the other machines in the other datacenters just won’t be used
|
2018-03-30 19:26:22 -07:00 |
|
Evan Tschannen
|
579ba58930
|
pop old tags only looks are recovered tags, and checks if they are still being used
|
2018-03-30 19:08:01 -07:00 |
|
Evan Tschannen
|
8352b93f48
|
fix: do not reuse tags that are still in historyTags, pop historyTags past epochEnd to allow tlogs to finish recovery
fix: peekLocal did not properly respect end
fix: the storage server added to the end of the history vector instead of the beginning
|
2018-03-30 17:39:45 -07:00 |
|
Evan Tschannen
|
43cb63df25
|
fix: the collectTags bool was set incorrectly
|
2018-03-29 18:19:29 -07:00 |