69711a107b
fix: because of forced recovery, 0 log router tags does not mean we are a special tlog set
2018-10-02 17:45:11 -07:00
e7e1c634e0
fix: we need to restart the peek cursor when the known committed version becomes available
2018-10-02 17:44:14 -07:00
a92fc911ac
do not spin on a failed storage server recruitment
2018-10-02 17:31:07 -07:00
dd9c6856e1
fix: forced recovery is not safe to send multiple times
2018-10-02 17:29:39 -07:00
15ce215c1b
fix: parallel peek requests leaked memory
2018-10-02 17:28:39 -07:00
bd6b743a81
fix: the storage server must always keep MAX_READ_TRANSACTION_LIFE_VERSIONS of history in memory, because forced recovery could roll back an epoch end.
...
fix: rollbacks were triggered unnecessarily
2018-09-28 16:04:59 -07:00
3fdf72c626
fix: we need to force recovery if the master is still attempting to read the txs tag
2018-09-28 13:33:33 -07:00
59335aa757
fix: the latest generation of remote transaction logs might has less data the a previous generation, because they take over at known committed version. Detect this case and end at the version that has the most data
2018-09-28 12:25:27 -07:00
c577840020
fix: forced recovery should remove all references to the old primary tlogs in all generations of logs to help the peek logic avoid attempting to read from them
2018-09-28 12:23:09 -07:00
05e7f08b26
added a peek method which will attempt to read the txsTag from the local region as much as possible
2018-09-28 12:21:08 -07:00
a24eadd73a
fix: for remote logs, their known committed version cannot be set to 1, because they can be used when their durable version is 0, leading to a known committed version being greater than a queue committed version
2018-09-28 12:17:21 -07:00
e64c55dce0
fix: data distribution would use the wrong priority sometimes when fixing an incomplete movement, this lead to the cluster thinking the data was replicated in all regions before it actually was
2018-09-28 12:15:23 -07:00
b1fe069165
fix: during forced recovery logs can be removed from the logSystemConfig. We need to avoid killing the removed logs as unneeded until we actually complete the recovery
2018-09-28 12:13:46 -07:00
22e6afbb18
fix: the cluster controller did not pass in its own locality when creating its database object, therefore it was not using locality aware load balancing
2018-09-28 12:12:06 -07:00
b560b94ebc
fix: do not force a recovery if the master was already in the other region (and therefore already recovered)
...
fix: reboot the remaining DC, because any storage server rejoins that were rolled back will cause that server to be unusable
2018-09-28 12:10:04 -07:00
77e2fb787e
Merge branch 'release-6.0' into feature-fix-forced-recovery
2018-09-21 14:55:37 -07:00
3f86905ea7
fix: restore did not take into account that the end version of a log file does not exist in that file. This resulted in restores done at the same version a snapshot completes to not apply the mutations at that final version.
2018-09-21 11:48:28 -07:00
6b6d7a087d
The cluster controller should never consider itself as failed (that will be handled by the coordinators)
...
Simplified the check that the cluster controller is excluded
2018-09-20 17:01:11 -07:00
31d0b0315f
fix: tlog spill policy would spill everything when it wanted to spill nothing
...
use a flow lock to protect updatePersistData and initPersistentState from committing simultaneously
2018-09-20 15:33:38 -07:00
03728db99b
do not trigger better master exists if the cluster controller is excluded, since the master will change anyways once the cluster controller is moved
2018-09-19 18:28:24 -07:00
42a67efb0c
the cluster controller should prefer to be located on a transaction class machine over a storage server class machine
2018-09-19 18:04:59 -07:00
861c8aa675
consider server health when building subsets of emergency teams
2018-09-19 17:57:01 -07:00
702d018882
fix: we cannot use count on an async map, because someone waiting onChange for an item will cause it to exist in the map before it is set
2018-09-19 16:11:57 -07:00
6d18193b3a
fix: team->setHealthy was not being called correctly on initially unhealthy teams
2018-09-19 14:48:07 -07:00
270b1b24a6
fix: we have to use durableKnownCommittedVersion, because the is the true lower bound on the recovery version of the remote logs
...
fixed a compiler error
2018-09-18 16:29:03 -07:00
200e65fe61
added a workload which tests killing an entire region, and recovering from the failure with data loss.
...
fix: we cannot pop the txs tag from remote logs until they have a full copy of the txnStateStore
fix: we have to modify all of history, we cannot stop after finding a local remote
2018-09-17 18:32:39 -07:00
16018ee36d
Merge pull request #770 from bnamasivayam/release-6.0
...
When the cluster is configured from fearless setup to usable_regions=…
2018-09-13 10:36:17 -07:00
d622cb1f6e
When the cluster is configured from fearless setup to usable_regions=1, master goes into a loop changing team priority . Fix this issue.
2018-09-12 18:29:49 -07:00
0c82edc5d9
Merge pull request #766 from alecgrieser/post-release-cleanup-6.0.11
...
Post release cleanup
2018-09-11 17:15:33 -07:00
20331e32d3
update installer WIX GUID following release
2018-09-11 17:05:09 -07:00
27125d9d65
update versions target to 6.0.12
2018-09-11 17:05:09 -07:00
700d558e5a
Merge pull request #764 from satherton/cleanup-status-refactor
...
Status refactor cleanup
2018-09-11 10:55:20 -07:00
b47902584b
Added comparison of serialization speed of json_spirit and JsonBuilder to JsonBuilder perf test.
2018-09-10 19:54:27 -07:00
5b8971ef18
Rename JsonString to JsonBuilder
2018-09-10 19:01:24 -07:00
5c23f91c59
Simplified templates that were used to forward rvalue references to JsonBuilder write() functions when possible as the current design does not require or benefit from this.
2018-09-10 18:48:30 -07:00
9e8c1941ec
Merge pull request #763 from alecgrieser/000761-getuserversion-order-of-operations
...
Fix issue with Versionstamp::getUserVersion on larger user versions
2018-09-10 18:21:54 -07:00
5128d9772d
cleverly use parentheses to ensure that the bitwise operation happens before the shift
...
This was casuing getUserVersion to return incorrect results if the user version was greater than 0x00FF.
Fixes #761
2018-09-10 18:18:13 -07:00
79f1b8963c
Merge pull request #762 from etschannen/release-6.0
...
prepare for 6.0.11
2018-09-10 18:10:57 -07:00
4c8a2df22b
updated downloads for 6.0.11
2018-09-10 18:08:39 -07:00
0dbb0244ff
updated wxs UID for 6.0.11
2018-09-10 18:04:20 -07:00
23a979a095
Merge pull request #758 from etschannen/release-6.0
...
fix: status would generate invalid json
2018-09-10 17:47:58 -07:00
5fff960e5d
updated release notes for 6.0.11
2018-09-10 17:45:01 -07:00
fb0d1d12a3
fix: reserve was not called correctly
2018-09-10 15:20:41 -07:00
02d7b02036
Moved readJSONStrictly to a client cpp to avoid duplicate definitions at link time.
2018-09-10 14:35:04 -07:00
8ed8f290c4
Merge branch 'release-6.0' of github.com:etschannen/foundationdb into evan-6
2018-09-10 13:28:49 -07:00
4753172991
Added support for NaN and infinity in JsonBuilder in double and ascii number interfaces.
2018-09-10 13:28:31 -07:00
63271072e8
fixed linker problem
2018-09-10 12:59:22 -07:00
9febc01106
In simulation, strict JSON parser will print bad input to stdout (not std since whether it's an error depends on context).
2018-09-10 12:07:03 -07:00
6a04fa0ede
In simulation, deserializing a StatusReply now does a strict JSON parse.
2018-09-10 11:13:41 -07:00
ce38ddbd4b
added tests with the + sign
2018-09-10 10:52:15 -07:00
Evan Tschannen