Commit Graph

49558 Commits (docker_plugin_improvements)

Author SHA1 Message Date
John Shahid aa61caaaa5 fix #370. Filtering and limit should be enforced on the shards 2014-03-31 18:55:28 -04:00
John Shahid e4335cf802 fix the shard boundaries test 2014-03-31 18:55:28 -04:00
John Shahid 3f543ead81 remove unused code 2014-03-31 18:55:28 -04:00
John Shahid 1d4d323e45 remove an unused func 2014-03-31 18:55:28 -04:00
Todd Persen 0769ff996e Add tests to verify that cluster admins can issue deletes. 2014-03-31 16:42:13 -04:00
Todd Persen e379a2fa65 Fix #381. Cluster admins should be able to issue deletes. 2014-03-31 16:28:03 -04:00
John Shahid 72fba5929a fix #378. Index should return immediately if there are no new requests 2014-03-31 14:31:04 -04:00
John Shahid 491924da94 add google-pprof cpu profiling
It currently doesn't play nice with go so it's disabled for now
2014-03-31 12:54:02 -04:00
Ben Johnson c6bdf639c2 Merge pull request #209 from tsenart/master
util: Handle all errors in writeFileSync
2014-03-29 14:37:47 -06:00
John Shahid d1e15fa7e3 update the changelog 2014-03-28 15:14:59 -04:00
John Shahid 92a7d37042 oops, fix a bug introduced in 1772fe9. Tests FTW 2014-03-28 13:50:44 -04:00
John Shahid 36582ec9ed update the changelog 2014-03-28 13:32:14 -04:00
John Shahid 1772fe9e7c fix #360. Store and recover continuous queries on startup 2014-03-28 13:27:31 -04:00
John Shahid c8f88c5bb5 use the ContinuousQuery structure instead of an interface 2014-03-28 13:00:19 -04:00
John Shahid 21f089d334 increase the timeout 2014-03-28 12:47:12 -04:00
John Shahid c5e2e2c171 don't close the server if it hasn't started yet 2014-03-28 12:47:03 -04:00
John Shahid 47bbb67de1 change the default log level to info 2014-03-28 12:40:03 -04:00
John Shahid 5c38e52a96 update the changelog 2014-03-28 12:39:05 -04:00
John Shahid e2490a245b fix #371. Seris index isn't deleted when the series is dropped 2014-03-28 12:38:35 -04:00
John Shahid d37d89d52f update the changelog 2014-03-28 11:54:46 -04:00
John Shahid 1567c69ad1 fix #331. Allow negative time values 2014-03-28 11:54:11 -04:00
John Shahid 99f6dc74bc update the changelog 2014-03-28 11:02:39 -04:00
John Shahid 6b890e91dc fix #355. Return an error on invalid time strings 2014-03-28 10:57:58 -04:00
John Shahid 40e7d08cc8 panic on error 2014-03-28 10:57:58 -04:00
John Shahid 33414d7c5e print the number of columns as well 2014-03-28 10:57:58 -04:00
John Shahid fb2ece556e panic if the WAL cannot append the request to disk 2014-03-28 10:57:58 -04:00
John Shahid fdd0118ac5 fix the shutdown logic when we're not profiling 2014-03-28 10:57:58 -04:00
Tomás Senart 2ab0a83793 util: Handle all errors in writeFileSync
Before this patch, errors returned by f.Write would be swallowed and
replaced by io.ErrShortWrite if the number of bytes written was less than the
number of bytes to be written.

etcd was panicking on startup because it had no space to write conf.tmp to its
data directory and the panic string was "short write" which was not very
informative of the real problem.

After applying this patch, we got a nice error message in our panic:
"panic: write /data/etcd/conf.tmp: no space left on device"
2014-03-28 15:03:08 +01:00
John Shahid b8dbac8cff renaming 2014-03-27 11:50:35 -04:00
John Shahid 49ee059e9a fix #367. Influxdb won't start if the api port is commented out 2014-03-27 11:50:35 -04:00
John Shahid bf883e8f9d update the changelog 2014-03-26 13:13:29 -04:00
John Shahid 7654856345 Merge pull request #342 from influxdb/fix-342-delete-bug
Deletes don't seem to actually delete the underlying data
2014-03-26 13:12:07 -04:00
John Shahid 87f49644de test dropping database as well 2014-03-26 12:59:22 -04:00
John Shahid 56866841aa fix #342. Make sure we create a checkpoint before deleting data
Deleting data without creating a wal bookmark will cause the log to
replay some requests that where written before the delete happened,
causing the data to be resurrected. This change makes sure we create a
bookmark right before we delete the data to make sure that any write
before the delete will not be replayed.
2014-03-26 12:59:22 -04:00
John Shahid 92463617f9 Add an api to create a wal checkpoint (i.e. bookmark and index) 2014-03-26 12:59:22 -04:00
John Shahid 089cb6fc13 add a tool to fix the index of a raft log file 2014-03-26 12:50:20 -04:00
John Shahid 10e1e0e3de formatting 2014-03-26 12:50:20 -04:00
John Shahid 7bcdc9cb6a Merge pull request #6 from kuba--/master
Fix typo in 'password'
2014-03-25 16:39:10 -04:00
Kuba Podgorski 1ff6b651b8 Fix typo 2014-03-25 13:04:31 -07:00
John Shahid 9ee31400ba ignore more files 2014-03-25 14:22:21 -04:00
John Shahid a15a2a9dc6 update the changelog 2014-03-25 13:20:16 -04:00
John Shahid af7455d256 checkin the source code of goraft 2014-03-25 12:22:37 -04:00
John Shahid 648206a3d6 oops, fix a typo 2014-03-24 19:59:56 -04:00
John Shahid 2366a9e9c9 make different submdoule version doesn't cause the release to stop 2014-03-24 19:59:28 -04:00
John Shahid 71660f80e7 change the goraft commit that we use 2014-03-24 19:58:30 -04:00
John Shahid 952ce0bfa1 Merge pull request #336 from influxdb/fix-336-raft-config
Make raft settings config options
2014-03-24 19:48:06 -04:00
John Shahid 549eff33de Merge pull request #359 from influxdb/fix-359-log-compaction
Influxdb isn't functional after restarting the entire cluster
2014-03-24 19:47:50 -04:00
John Shahid 7696d6e461 add goraft as a git submodule 2014-03-24 19:44:27 -04:00
John Shahid 9c22aab733 add a test to demonstrate a bug in the log compaction 2014-03-24 18:20:06 -04:00
John Shahid 9b1f92220f ignore more files 2014-03-24 16:59:54 -04:00