Commit Graph

344 Commits (62cb3a1e9b53f3ed6984574af2d6997c39c0cc0a)

Author SHA1 Message Date
Jason Wilder 62cb3a1e9b Merge pull request #5057 from influxdb/jw-5046
Fix leaking TSM files when compacting
2015-12-08 13:11:46 -07:00
Jason Wilder 3543917a74 Avoid allocating strings during search 2015-12-08 13:02:17 -07:00
Jason Wilder 99c313ddae Fix leaking TSM files when compacting
The files being read were not closed after the compaction ran causing
them to leak.

Fixes #5046
2015-12-08 12:55:30 -07:00
Jason Wilder 9d82e24ca0 Fix performance of dropping large number of keys 2015-12-08 10:47:06 -07:00
Jason Wilder f245b44afa Set full compaction duration option on planner
Was set on engine and not planner so it was always 0.
2015-12-08 09:56:36 -07:00
Jason Wilder d32aeb2535 Merge pull request #5031 from influxdb/jw-mintime
Dedupe points at query time if there are overlapping blocks
2015-12-07 21:28:29 -07:00
Jason Wilder 87892d79da Dedupe points at query time if there are overlapping blocks 2015-12-07 21:10:10 -07:00
Fazal Majid bb386219f4 ran go fmt on mmap_solaris.go #4787 2015-12-07 17:41:26 -08:00
Fazal Majid 0f889a77d1 fix tsm1 for Solaris #4787, passes unit tests now 2015-12-07 17:14:26 -08:00
Jason Wilder a2583d2be1 Reduce lock contention when planning TSM queries 2015-12-07 15:42:36 -07:00
Jason Wilder 4da20c49e9 Optimize TSM file scanning for time queries
Move the index locations planning to be lazily created after the first
seek when we know what time and direction we're searching for.  This
allows files and blocks to be skip before having to scan the files index.

This improves queries times with time filters wherne there are many TSM
files on disk.
2015-12-07 15:42:36 -07:00
Paul Dix 93d6afec97 Merge pull request #5019 from influxdb/jw-mintime
Remove min time from TSM blocks
2015-12-07 15:00:12 -05:00
Paul Dix 8096c6b845 Update TSM, address PR #5011 comments
* Moved TSM file extension to a constant
* Fixed typos
* Changed group.size() back to being a uint64 since it can have multiple files up to 4GB each.
2015-12-07 14:47:17 -05:00
Paul Dix 820b0d31d6 Update TSM to delete from the WAL/cache
* Update cache loader to delete entries from cache
* Add cache.Delete()
* Update delete to look at keys in the Cache in addition to the FileStore
* Update cache compaction to never happen if the cache is empty
2015-12-07 14:35:48 -05:00
Jason Wilder cf341eaa6a Remove MinTime from blocks
MinTime is not in the index for each block so storing it in the block
header is redundant.  The encodings also store it in their header so
we are actually storing it 3 times.

Removing this is an incompatible change with the current tsm1 file format.
2015-12-07 11:26:58 -07:00
Paul Dix 440a8a8a1f Change all TSM file sizes to uint32 2015-12-07 10:12:24 -05:00
Paul Dix 937233d988 Update TSM compaction planning logic
* Update Plan to do a full compaction if cold for writes
* Remove MaxFileSize as a config variable from Compactor. Should be a set constant
* Update Plan to keep track of if the last check was fully compacted so we can skip future planning calls
* Update compact min file count to 3 so that compactions run more frequently
2015-12-07 08:26:30 -05:00
Paul Dix 1bee7d1512 Update TSM, remove old version, add config
* remove rolloverTSMFileSize constant that is no longer used
* remove the maxGenerationFileCount since it is no longer a limitation that's necessary with the new compaction scheme. We no longer read WAL segments as part of the compaction so memory is only used as we read in each individual key
* remove minFileCount and switch to a user configurable variable
* remove the mutex from WALSegmentWriter. There's never more than one open in the WAL at one time and it's not exported through any function so the lock on the WAL should be used. This simplified keeping track of the last write time and removed a bunch of unnecessary locks.
* update WALSegmentWriter.Write to take the compressed bytes so that encoding and compression can occur before the call to write (while we don't hold the WAL lock)
* remove a bunch of unnecessary locking in WAL.writeToLog
* Add check for TSM file magic number and vesion
* Remove old tsm, log, and unused cursor code
* Remove references to tsm1dev everywhere except in the inspector
* Clean up config options for compaction and snapshotting
* Remove old TSM configuration options
* Update the config.sample.toml with TSM options
* Update WAL compact to force if it has been cold for writes for a configurable period of time (1h by default)
2015-12-06 18:50:39 -05:00
Philip O'Toole 6e88547a5e Support shutting down engine goroutines
This was causing races in the code, when the cache was being reloaded,
because back-to-back open-and-closing of the engine during testing left
goroutines running. With this change the engine is completely shutdown
when Close() is called on it.
2015-12-06 09:16:38 -08:00
Philip O'Toole 0d0b919144 Integrate CacheLoader with tsm2 engine 2015-12-05 22:13:57 -08:00
Philip O'Toole fe7b3ad134 Add CacheLoader
The CacheLoader loads a given cache from a slice of segment files.
2015-12-05 22:13:57 -08:00
Philip O'Toole 4b5fb8db72 WALSegmentReader counts bytes read without error 2015-12-05 22:13:57 -08:00
Philip O'Toole c67831bc79 Remove double-checking of error when reading WAL 2015-12-05 22:13:57 -08:00
Paul Dix 40e606cb14 Merge pull request #5003 from influxdb/jw-compaction
Update compaction planning
2015-12-05 16:49:54 -05:00
Jason Wilder 33a33e6a23 Fix 32bit int overflow of constant value 2015-12-05 13:09:18 -07:00
Jason Wilder 41b24995a7 Compcation fixes 2015-12-05 12:19:28 -07:00
Philip O'Toole 7296de1fac Merge pull request #4999 from influxdb/cache_sort
Always copy the Cache values for query and merge with snapshot
2015-12-05 08:15:13 -08:00
Philip O'Toole 1b12ff9c1c Only take write-lock for Values when necessary 2015-12-05 08:06:01 -08:00
Jason Wilder 6592615958 Updated compaction strategy
This changes compacting files to merge sequences of files in lower generations
up to later generations
2015-12-04 23:30:39 -07:00
Philip O'Toole 789ab10658 Merge hot cache values with snapshots
This change starts by building the sequence of entries, which also
allows the required size of destination buffer to be calculated. Then
the buffer is allocated up-front in 1 call.

Each snapshot and hot value-set is appended to the buffer. If ordering
is violated at anytime, set the 'needSort' flag. Sorting, if necessary,
is performed just before returning the data.
2015-12-04 20:58:02 -08:00
Philip O'Toole 859877fd09 Move all sort logic to entry type 2015-12-04 20:21:16 -08:00
Philip O'Toole 6e91679fab Always copy the Cache values for query 2015-12-04 15:37:45 -08:00
Paul Dix 9637446ba9 Merge pull request #4990 from influxdb/pd-loadmetadata-wal
Update TSM engine, WAL and encoding
2015-12-04 18:21:47 -05:00
Paul Dix 33506e4d3e Update TSM cache and engine LoadMetadataIndex 2015-12-04 16:40:01 -05:00
Paul Dix b0f3dcc8cc Update TSM metadata loading and write snapshot
* Update WriteSnapshot to always call synchronously
* Update LoadMetadataIndex to load WAL metadata from the cache
2015-12-04 16:03:17 -05:00
Jason Wilder 357b88c439 Increment sequence of max generation when compaction files 2015-12-04 13:46:28 -07:00
Jason Wilder 52bec1f7f6 Change TSM file naming to generation-sequence.tsm 2015-12-04 11:51:33 -07:00
Jason Wilder 479469994a Optimize FileStats calls
FileStats called frequently during compaction planning was too expensive because
they were cleared out every time a file replaced causing them all to be reloaded.
Insted, we grab the stats that are already maintained by the files themselves from
the files when needed.
2015-12-04 11:16:39 -07:00
Jason Wilder 70710df910 Fix typo 2015-12-04 10:02:59 -07:00
Jason Wilder c7e37766e7 Avoid repetitive index searches when iterating over cursors
First pass at TSM cursor iteration ended up searching the file indexes
too frequently and hurt performance.  This changes that to search it once
and then have the cursor hold onto the block locations to seek
to.  Doubles the query performance from the first iteration, but still a lot
of room for improvement.
2015-12-04 10:02:59 -07:00
Jason Wilder 4b7cc6720a Merge pull request #4983 from influxdb/jw-tsm-deletes2
Implement delete series/measurement
2015-12-04 10:02:11 -07:00
Jason Wilder c54a3da0ca Implement delete series/measurement 2015-12-04 09:10:26 -07:00
Paul Dix eafb703afc Update TSM engine, WAL and encoding
* Add InfluxQLType to Values to map the TSM type to InfluxQL
* Fix bug in WAL where close wouldn't nil out the currentSegment after closing it
* Export writeSnapshot to be used in tests, add argument to run it async or not
* Update reloadCache to load temporary metadata information in the engine
* Update LoadMetadataIndex to use the temp WAL metadata information
2015-12-04 11:09:39 -05:00
Philip O'Toole 89a3490cae Merge pull request #4989 from influxdb/cache_rename
Fix comment and remove snapshot stutter
2015-12-04 07:43:26 -08:00
Philip O'Toole f939e49f0f Fix comment and remove snapshot stutter 2015-12-04 07:29:58 -08:00
Paul Dix b7bae53405 Merge pull request #4980 from influxdb/cursor_desc
Fix descending cache cursor
2015-12-04 07:02:13 -05:00
Philip O'Toole 2d79d7e35f Fix descending cache cursor 2015-12-03 14:34:29 -08:00
Jason Wilder 66c9ef862e Fix regressions
Something broke with writing to the WAL now that compactions are running
concurrently.  There was also a performance problem with Next/Prev doing
twice as many searches as necessary.
2015-12-03 14:25:03 -07:00
Jason Wilder adf5c5b223 Replace Next/Prev with Scan 2015-12-03 12:39:13 -07:00
Jason Wilder 193a36eeb6 Fix code review comments 2015-12-03 12:39:13 -07:00