Influxdata
/
influxdb
mirror of https://github.com/influxdata/influxdb.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
9,171 Commits
243 Branches
453 Tags
345 MiB
Commit Graph

9171 Commits (771253256b1e61f6a149e7ae37a9e5bff0400233)

Author SHA1 Message Date
INADA Naoki 771253256b FloatValue uses unixnano instead of time.Time 2016-02-03 09:57:00 +09:00
INADA Naoki 898babf616 add float bench 2016-02-03 03:12:16 +09:00
David Norton 5a9953eec1 Merge pull request #5511 from seiflotfy/master 
Initialize MetaClient before setting it up as a data and/or meta node
 2016-02-02 12:00:49 -05:00
Seif Lotfy c3fb710871 Initialize MetaClient before setting it up as a data and/or meta node 
Take out MetaClient initialization from initializeDataNode and invoke it
in (server.)Open after MetaService is opened.
Fixes #5479
 2016-02-02 17:39:29 +01:00
Edd Robinson 71ccbe5712 Merge pull request #5495 from influxdata/er-exit-code 
Ensure non-interactive mode returns appropriate exit code
 2016-02-02 15:35:37 +00:00
Edward Robinson b8e42cdd3a Ensure non-interactive mode returns appropriate exit code 
Fixes #5475.
 2016-02-02 15:23:16 +00:00
Cory LaNou fd85ae5b7f Merge pull request #5498 from influxdata/fix-vet 
Fix vet error
 2016-02-01 15:45:19 -06:00
Cory LaNou a171089fce fix vet error 2016-02-01 15:33:54 -06:00
joelegasse 07d51c8c88 Merge pull request #5493 from influxdata/jl-issue-5478 
Update first() and last() to handle varying types
 2016-02-01 15:46:24 -05:00
Joe LeGasse 4b642fb365 Added tests for tsdb.greaterThan() 
Also updated style to match "Effective Go" recommendations.
 2016-02-01 15:11:17 -05:00
Cory LaNou 953584aa31 Merge pull request #5494 from influxdata/ignore-join-urls 
Ignore join arguments on restart
 2016-02-01 12:28:44 -06:00
Cory LaNou f8b1aa69a1 ignore join arguments on restart 2016-02-01 12:47:15 -05:00
Joe LeGasse 2bd85b3c9e Update first() and last() to handle varying types 2016-02-01 11:38:08 -05:00
Jason Wilder 60df13fb56 Merge pull request #5476 from influxdata/jw-fixes 
Fix two panics in the WAL
 2016-01-29 09:08:35 -07:00
Jason Wilder 924275b337 Fix panic preventing wal file truncation 
Fixes #5455
 2016-01-28 21:50:51 -07:00
Jason Wilder 01193668cf Fix nil pointer panic when dropping collectd points 
Fixes #5449
 2016-01-28 21:36:59 -07:00
joelegasse 834744a4c3 Merge pull request #5472 from influxdata/jl-backup-dir 
Update influx_tsm to use a backup directory
 2016-01-28 13:35:57 -05:00
Joe LeGasse d2816dbb34 Updated CHANGELOG 2016-01-28 13:11:26 -05:00
Joe LeGasse 8f3131b97d Update influx_tsm to use a backup directory 2016-01-28 12:36:34 -05:00
Cory LaNou 797126c570 Merge pull request #5447 from influxdata/fix-5426 
Allow for node upgrade
 2016-01-28 11:21:56 -06:00
Cory LaNou 0f6c75ab7d make tests pass 2016-01-28 11:03:44 -06:00
Cory LaNou 23de1b15aa error out upgrading a cluster 2016-01-28 10:05:56 -06:00
Cory LaNou 51f6c64134 ake new and upgrade behavior the same for meta/data node numbering 2016-01-28 10:05:56 -06:00
Cory LaNou 53323737b2 no longer need nil check 2016-01-28 10:05:54 -06:00
Cory LaNou d70b694d7d fix misc meta statup bugs 2016-01-28 10:05:53 -06:00
Cory LaNou 31c2e7012a allow for node upgrade 2016-01-28 10:05:53 -06:00
joelegasse 9ab5de0a34 Merge pull request #5454 from joelegasse/influx_tsm 
TSM conversion tool improvements
 2016-01-28 10:38:47 -05:00
Joe LeGasse 482772997e Updated tests for 'influx_tsm' 
Also changed some things to fix failing tests on circleCI, and
removed old TODO item
 2016-01-28 09:34:00 -05:00
Jason Wilder d771eb72f7 Merge pull request #5467 from influxdata/jw-backup-node 
Backup node.json with metastore backup
 2016-01-27 23:06:14 -07:00
Jason Wilder 716714364a Backup node.json with metastore backup 2016-01-27 17:39:54 -07:00
Joe LeGasse 908259340b Improvements to influx_tsm 
- Improve logging and status updates
- Added debug http endpoint
- Properly resume/skip backed up files
 2016-01-27 16:13:23 -08:00
Todd Persen 0328ac1a7e Fix typo in README.md 2016-01-27 15:16:37 -08:00
Todd Persen 265e3aff8c Update CHANGELOG.md 2016-01-27 15:15:16 -08:00
Todd Persen 3723680747 Merge pull request #5460 from sczk/issue-5436 
Prevent exponential growth in ~/.influx_history
 2016-01-27 15:12:31 -08:00
Jason Wilder 9528c3ea70 Merge pull request #5465 from influxdata/jw-remote-writes 
Optimize remote writes
 2016-01-27 15:47:02 -07:00
Ben Johnson bc312ceb14 Merge pull request #5463 from benbjohnson/tsm1-perf 
tsm1 query performance improvements
 2016-01-27 15:22:05 -07:00
Jason Wilder 1d165d38a9 Optimize Cache entry.add 
This reduces some of the lock contention when writing to the cache.
When a new entry is created, it avoids an allocation.  It also skips
a check to see if we need to sorted if we already know it needs to sorted.
 2016-01-27 14:26:42 -07:00
Jason Wilder d54f930c2d Don't parse points twice when receiving remote writes 
The monitoring stats were causing points to be parsed twice create
more cpu time just parsing points.
 2016-01-27 14:24:56 -07:00
Jason Wilder 47c5ade858 Use faster point parsing for remote writes 
Parsing the line protocol again on the receiving side of the remote
write consumes a lot cpu.  This uses a different marshaling format
that is much faster to parse after we already parsed the point on
the write side.
 2016-01-27 14:24:09 -07:00
Ben Johnson 98baf078d0 tsm1 query performance improvements 2016-01-27 13:42:32 -07:00
Joe LeGasse 4f89c15bd3 Replaced more log.Print();os.Exit(1) with log.Fatal() 2016-01-27 07:25:46 -08:00
Adam Svoboda 40e04d89fc Prevent exponential growth in ~/.influx_history 
The history file is cleared before WriteHistory is called after each
command/exit() to prevent exponential file growth.

This commit addresses issue #5436, please see PR for full explanation.
 2016-01-26 20:53:41 -06:00
Todd Persen e5fa969306 Update CHANGELOG.md 2016-01-26 18:30:05 -08:00
Todd Persen 308323f48f Merge pull request #5459 from influxdata/tp-ping-to-status 
Remove MetaClient.Ping from `/ping` and move it to `/status`
 2016-01-26 18:26:37 -08:00
Jason Wilder 1255bb9cb0 Merge pull request #5458 from influxdata/jw-write-pool 
Use a bounded pool for remote writes
 2016-01-26 19:13:33 -07:00
Todd Persen 66e6375973 Move status request metrics to their own label 2016-01-26 18:10:02 -08:00
Todd Persen 06e91dfca1 Remove MetaClient.Ping from `/ping` and move it to `/status` 2016-01-26 17:58:44 -08:00
Jason Wilder 5abdb42a7d Use a bounded pool for remote writes 
Under highly conncurrent write load, the coordinating node would
create a connection to any other node that is part of the replica
group.  Since each connection can be expensive, OOM sitations could
occur because there was no bounds on the number of new connections
that would be created.  If writes on a remote node were slow, connections
could pile up an exacerbate the problem.

This switches the pool to be bounded and has a checkout that is blocking
with a timeout.  If a connection is available, it's returned immediately.
If the pool still has room for more connections, it will create one if needed.
Otherwise, the call will block until a connection becomes available or
the timeout expires.  In the case of a timeout, it is propogated back up
to the PointsWriter that determine what do return to the client.
 2016-01-26 17:08:36 -07:00
Jason Wilder 697f48b4e6 Merge pull request #5445 from influxdata/jw-oom 
Reduce lock contention in engine.WritePoints
 2016-01-26 11:30:31 -07:00
Joe LeGasse cdde2959af Limit parallelism for 'influx_tsm -parallel' 2016-01-26 09:11:09 -08:00