Commit graph

7615 commits

Author SHA1 Message Date
tgruben
7740036717
Merge pull request #1647 from tgruben/core-527
[CORE-527] multi-node to single backup check
2021-07-01 15:17:19 -05:00
Todd Gruben
4739ebf79e multi-node to single backup check 2021-07-01 14:02:16 -05:00
tgruben
69ef605f93
Merge pull request #1646 from tgruben/ci-backup
[CORE-629] CI test for backup restore
2021-07-01 09:55:50 -05:00
Todd Gruben
2f4f75a6cb renamed docker compose image 2021-07-01 08:56:41 -05:00
Todd Gruben
039ebc7ea6 trying docker login 2021-06-30 16:56:57 -05:00
Todd Gruben
74c03b1a33 needed executor 2021-06-30 15:02:22 -05:00
Todd Gruben
bac9ac047c Merge branch 'ci-backup' of github.com:tgruben/privilosa into ci-backup 2021-06-30 14:59:05 -05:00
Todd Gruben
87d2b513d8 corrected ci config 2021-06-30 14:58:39 -05:00
tgruben
72b2896178
Merge branch 'master' into ci-backup 2021-06-30 14:55:59 -05:00
Todd Gruben
3f8047778c CI test for backup restore 2021-06-30 14:52:15 -05:00
Nia
15c6023806
Merge pull request #1643 from seebs/keepInTranslation
[CORE-727] separate backup of index data and index translation keys
2021-06-30 13:39:05 -04:00
Nia
bc01e55280
Merge branch 'master' into keepInTranslation 2021-06-30 13:11:58 -04:00
Nia
37f291bb5b
Merge pull request #1644 from niaow/backup-dir-fsync
[CORE-726] fsync all directories after completing a backup
2021-06-30 13:11:49 -04:00
Seebs
90c424a4d9 separate backup of index data and index translation keys
It's not enough to back up each index's translation keys after
backing up that index's data; we also have to back them up after
backing up any index data from indexes which have foreign key
references to that index. So we do the per-index passes separately.

Since the individual backup data files are being created separately,
the expected output is unchanged for a quiescent database, the only
difference is that the amount of translation info which might be
newer than the data stored for shards is potentially increased.
2021-06-30 11:46:47 -05:00
Nia Weiss
779e27dcf6
fsync all directories after completing a backup
Previously the backup tool only fsync'ed the files.
Since the directories were not synced, it was possible for the references to be lost.
Now we sync the entire output directory tree and its parent.
2021-06-30 12:39:17 -04:00
tgruben
a8161923de
Merge pull request #1642 from tgruben/index-delete-shard
delete index missed index/shard removal [Bug]
2021-06-25 12:10:08 -05:00
tgruben
b15b11462a
Merge branch 'master' into index-delete-shard 2021-06-25 11:26:31 -05:00
Todd Gruben
ccda5f46d5 delete index missed index/shard removal 2021-06-25 11:24:28 -05:00
Mahesh Arumugam
5187d3f862
Merge pull request #1641 from molecula/ma/cloud-119
CLOUD-119 CORE-653 Fix percentile query
2021-06-24 17:45:07 -07:00
Mahesh Arumugam
357caf68c3 Fix percentile query: field is mandatory (should not crash), fieldnames can be unquoted 2021-06-24 15:10:05 -07:00
seebs
3a678c9ce0
Merge pull request #1639 from molecula/ciFailures
address CI failures due to timeouts
2021-06-24 11:11:10 -05:00
seebs
965a325f9d
Merge branch 'master' into ciFailures 2021-06-24 09:48:06 -05:00
tgruben
69dc3593a5
Merge pull request #1640 from tgruben/chksum
chksum sub-command for easier data validation
2021-06-23 16:38:44 -05:00
Todd Gruben
44281fa5c2 add chksum command for easier data validation 2021-06-23 16:10:52 -05:00
Seebs
0e1d448baa explicitly revoke lease on shutdown
If we explicitly shut a node down, we don't want everyone else
thinking it's up for the next 5 seconds. Worse, in CI, we have random
long delays (10+ seconds) with no CPU activity at all, so we have to
set the TTL longer there. Which makes any test checking for responsive
detection of a node going down take even longer. So! We revoke
leases on our way down, and this makes the tests not take so
long, and allows us to have a reasonable timeout on the test, while
having a completely unreasonable HeartbeatTTL to make CI stop
breaking randomly.
2021-06-22 09:02:25 -05:00
Seebs
13dd357ddd bump heartbeatTTL to ludicrous value (60s) for testing
After continuing to see weird test failures, did some more careful testing,
discovered that CI can pause a machine entirely for up to 29 seconds
or so very rarely, and 5-10 seconds quite frequently, which causes
cascading heartbeat failures and so on. Remove those.
2021-06-22 09:02:25 -05:00
Seebs
5898c58a80 try to force TMPDIR to be the ram disk
CircleCI is providing a ramdisk as /mnt/ramdisk. Using the ramdisk
instead of local storage makes fsync operations essentially free,
which removes (some of) the frequent multi-second delays we see during
runs otherwise.
2021-06-22 09:02:25 -05:00
Seebs
01103b26f0 make etcd bootstrap timeout configurable
It turns out that it's desireable to be able to configure the bootstrap
timeout for etcd, because during startup, we end up delaying that long
(N-1) times in series during each cluster creation, which is pointless
when we're starting the whole cluster. Reduces test runtime by several
minutes.
2021-06-22 09:02:25 -05:00
Ben Johnson
e8a3c313ac
Merge pull request #1638 from molecula/parallelize-restore
CORE-640: Parallelize restore command
2021-06-21 15:07:58 -06:00
Todd Gruben
23ba86cbfc linter fix 2021-06-21 14:15:13 -05:00
Ben Johnson
dad2372a8c CORE-640: Parallelize restore command 2021-06-18 16:03:38 -06:00
Ben Johnson
e082f6bce0
Merge pull request #1637 from molecula/parallel-backup
CORE-639: Parallelize backup
2021-06-15 17:44:17 -06:00
Ben Johnson
3df1f2d094 CORE-639: Parallelize backup 2021-06-15 16:55:05 -06:00
Samir Patel
f5844a0eb3
Merge pull request #1633 from 54mir/reset-cache-on-schema
CORE-642 Reset Usage Cache If Outdated On Request
2021-06-15 13:01:58 -05:00
Samir Patel
92ff74c72d
Merge branch 'master' into reset-cache-on-schema 2021-06-15 10:32:43 -05:00
Samir Patel
6b45325ca7 expand comment on duty cycle 2021-06-15 09:37:42 -05:00
Samir Patel
fb11d7f656
Update api.go
Co-authored-by: Alan Bernstein <alanaaronbernstein@gmail.com>
2021-06-15 09:17:53 -05:00
Stephanie Yang
25bf4fdaa7
Merge pull request #1636 from molecula/feature/syang/diskusage-lastupdated
CLOUD-78: Add 'last updated' info to tables UI
2021-06-14 16:25:22 -05:00
Stephanie Yang
7c89eb8033 add cache refresh copy 2021-06-14 16:02:15 -05:00
Stephanie Yang
9c21f6f109 consistency 2021-06-14 15:49:20 -05:00
Stephanie Yang
b14811b4fe add UTC time tooltip when hovering ov er relative time 2021-06-14 15:47:09 -05:00
Stephanie Yang
f48d7998e1 add last updated info for disk usage to ui 2021-06-14 15:47:09 -05:00
Stephanie Yang
e4159a1927 fix UI blow up when index disk usage not available 2021-06-14 15:47:09 -05:00
Ben Johnson
ee3a7a845e
Merge pull request #1635 from molecula/backup-dir
CORE-638: Refactor backup/restore to use directory archive
2021-06-14 11:03:39 -06:00
Samir Patel
d83a502657 change comment 2021-06-14 11:55:23 -05:00
Samir Patel
bfc90d148d Merge branch 'reset-cache-on-schema' of github.com:54mir/pilosa into reset-cache-on-schema 2021-06-14 11:53:46 -05:00
Samir Patel
6bdf685471 change waitMultiplier to float64 2021-06-14 11:43:49 -05:00
Samir Patel
11207fc589
Update api.go
Co-authored-by: Alan Bernstein <alanaaronbernstein@gmail.com>
2021-06-14 11:12:12 -05:00
Samir Patel
8bf472bb75 add duty cycle config flag 2021-06-14 10:48:52 -05:00
Ben Johnson
73f0668fe1 Refactor backup/restore to use directory archive 2021-06-14 09:43:20 -06:00