wbtiger
/
foundationdb
mirror of https://github.com/apple/foundationdb.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
3,022 Commits 37 Branches 304 Tags 268 MiB
Commit Graph

3022 Commits

Author SHA1 Message Date
Stephen Atherton 7c1dc305cb Merge commit 'a72c8f5cb2e79a673abc0ed3d27ef1c51028fb13' into feature-redwood 2018-10-05 10:15:10 -07:00
Stephen Atherton 5d9cd9acdc Correctness test now has additional random reader which doesn't do verification but isn't stopped when the btree is closed. Fixed bug exposed by this where pager snapshots will still try to read pages after the pager has been shut down or even destroyed. Added new error type, shutdown_in_progress. 2018-10-04 23:46:37 -07:00
A.J. Beamon a72c8f5cb2
Merge pull request #811 from apple/release-6.0 
Merge release-6.0 into master
 2018-10-04 13:30:49 -07:00
A.J. Beamon d02b6bf211
Merge pull request #810 from apple/release-5.2 
Merge release-5.2 into release-6.0
 2018-10-04 13:30:26 -07:00
A.J. Beamon b3912b3918
Merge pull request #809 from apple/release-5.1 
Merge release-5.1 into release-5.2
 2018-10-04 13:29:58 -07:00
A.J. Beamon e7718724fb
Merge pull request #808 from ajbeamon/release-5.1 
Fix typo
 2018-10-04 13:29:27 -07:00
A.J. Beamon abaefbe294 Fix typo 2018-10-04 13:28:35 -07:00
A.J. Beamon fc0dbe7c66
Merge pull request #807 from apple/release-6.0 
Merge release-6.0 into master
 2018-10-04 13:26:32 -07:00
A.J. Beamon d7cf6a7759
Merge pull request #806 from apple/release-5.2 
Merge release-5.2 into release-6.0
 2018-10-04 13:25:13 -07:00
A.J. Beamon 1b7aac6f5a
Merge pull request #805 from apple/release-5.1 
Merge release-5.1 into release-5.2
 2018-10-04 13:24:32 -07:00
A.J. Beamon f215ae037a
Merge pull request #804 from apple/release-5.1 
Merge release-5.1 into release-5.2
 2018-10-04 13:23:16 -07:00
Evan Tschannen fc098586a1
Merge pull request #802 from etschannen/master 
Merge 6.0 into master
 2018-10-03 16:58:14 -07:00
Evan Tschannen 3922e477a5 Merge branch 'release-6.0' 
# Conflicts:
#	documentation/sphinx/source/release-notes.rst
#	fdbclient/ManagementAPI.actor.cpp
#	fdbserver/ClusterController.actor.cpp
#	fdbserver/DataDistribution.actor.cpp
#	fdbserver/LogSystemDiskQueueAdapter.actor.cpp
#	fdbserver/SimulatedCluster.actor.cpp
#	fdbserver/TLogServer.actor.cpp
 2018-10-03 16:57:18 -07:00
Evan Tschannen 9de55f362b
Merge pull request #793 from ajbeamon/add-new-storage-status-metrics 
Add new metrics for bytes queried, keys queried, mutation bytes, muta…
 2018-10-03 16:34:26 -07:00
Evan Tschannen 5b7ca7e4b1
Merge pull request #796 from brownleej/monitor-conf-system-key-range 
Adds a key range for storing changes to monitor conf files.
 2018-10-03 16:33:05 -07:00
Evan Tschannen a91f59209b
Merge pull request #799 from ajbeamon/release-5.1 
Add release notes for 5.0.8 to old release notes
 2018-10-03 16:32:28 -07:00
Evan Tschannen 598788f60b
Merge pull request #801 from etschannen/feature-fix-forced-recovery 
Fixed a number of problems with forced recoveries
 2018-10-03 16:32:03 -07:00
Evan Tschannen 4bec4927aa updated release notes for 6.0.12 2018-10-03 14:32:28 -07:00
Evan Tschannen 636420abee fix: if the disk queue adapter peek hangs for a while, switch to a peek from a different locality 2018-10-03 13:58:55 -07:00
Evan Tschannen 28545e0f8d multi cursors start a get more for the first 10 cursors to hide latency 2018-10-03 13:57:45 -07:00
Evan Tschannen aa51d69b2d fix: set peekLocality for upgraded tags 2018-10-03 13:54:59 -07:00
A.J. Beamon 87ff41006b
Merge pull request #800 from ajbeamon/master 
Fix API version in old release notes link
 2018-10-03 12:35:43 -07:00
A.J. Beamon efb081ad86 Fix API version in old release notes link 2018-10-03 12:33:35 -07:00
Stephen Atherton c5420218d0 Missed changes from previous checkin which stops freeing physical pages after a latest version logical page rewrite. 2018-10-03 12:10:07 -07:00
A.J. Beamon 54c240cbed Add release notes for 5.0.8 to old release notes 2018-10-03 08:51:47 -07:00
A.J. Beamon 02f409b235
Merge pull request #798 from ajbeamon/master 
Preliminary setup for 6.1.0 release notes
 2018-10-03 08:34:13 -07:00
A.J. Beamon 14a0427462 Preliminary setup for 6.1.0 release notes 2018-10-03 08:32:19 -07:00
Evan Tschannen c9f4109539 fix: add some additional time in the kill region workload to detect if we recovered successfully 2018-10-02 17:47:15 -07:00
Evan Tschannen cdaf5e1192 fix: forced recovery does not recover tags from any DC besides the surviving one 2018-10-02 17:46:22 -07:00
Evan Tschannen 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
Evan Tschannen e7e1c634e0 fix: we need to restart the peek cursor when the known committed version becomes available 2018-10-02 17:44:14 -07:00
Evan Tschannen a92fc911ac do not spin on a failed storage server recruitment 2018-10-02 17:31:07 -07:00
Evan Tschannen dd9c6856e1 fix: forced recovery is not safe to send multiple times 2018-10-02 17:29:39 -07:00
Evan Tschannen 15ce215c1b fix: parallel peek requests leaked memory 2018-10-02 17:28:39 -07:00
Stephen Atherton 4ad44e72dd Workaround for issue where calling sync on a cached file that doesn't exist yet causes an io_error after a simulated machine kill. Temporary disabled freeing physical pages on logical page replacement because it is a bug as currently implemented, it will reuse physical pages before the remap of its old logical page to a new physical page is durable. 2018-10-02 17:00:35 -07:00
Stephen Atherton 15c1f87681 Rewrite superpage writes to fix bugs involving logical ID reuse and to make the logic cleaner. Correctness test is stronger, with pipelined writes/commits during verification. 2018-10-01 16:51:57 -07:00
A.J. Beamon 84c2e3567f Fix keys queried to use the RowsQueried metric instead of BytesQueried. 2018-10-01 11:19:28 -07:00
John Brownlee 2beeadf8be Adds a key range for storing changes to monitor conf files. 2018-10-01 10:49:02 -07:00
A.J. Beamon a98fcf5972 Rename durable_lag to durability_lag 2018-10-01 09:58:49 -07:00
Evan Tschannen 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
Evan Tschannen 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
Evan Tschannen 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
Evan Tschannen 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
Evan Tschannen 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
Evan Tschannen 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
Evan Tschannen 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
Evan Tschannen 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
Evan Tschannen 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
Evan Tschannen 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
Stephen Atherton a9f467c502 Btree correctness verification of contents is now asynchronous. 2018-09-28 00:35:03 -07:00