Aliaksandr Valialkin
fdc5cfd838
lib/mergeset: reduce the maximum number of cached blocks, since there are reports on OOMs due to too big caches
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/189
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/195
2019-09-30 12:25:40 +03:00
Aliaksandr Valialkin
0fa6df94a2
lib/storage: optimize TSID comparison
2019-09-26 14:16:02 +03:00
Aliaksandr Valialkin
c39355921e
lib/storage: verify whether items are sorted in the end of call to mergeTagToMetricIDsRows
...
This should prevent from inverted index corruption if bug in mergeTagToMetricIDsRows is discovered.
2019-09-26 13:13:41 +03:00
Aliaksandr Valialkin
0db9fcedd5
lib/storage: properly match labels against regexp with (?i)
flag
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/161
2019-09-26 11:03:10 +03:00
Aliaksandr Valialkin
2444433d83
lib/storage: add missing break in removeDuplicateMetricIDs
2019-09-25 18:23:43 +03:00
Aliaksandr Valialkin
ea4c828bae
lib/storage: remove duplicate MetricIDs in tag->metricIDs
items before writing them into inverted index
2019-09-25 17:55:13 +03:00
Aliaksandr Valialkin
aebc45ad26
lib/{mergeset,storage}: do not cache inverted index blocks containing tag->metricIDs
items
...
This should reduce the amounts of used RAM during queries with filters over big number of time series.
2019-09-25 14:02:15 +03:00
Aliaksandr Valialkin
2cb811b42f
lib/uint64set: optimize Set.AppendTo
2019-09-25 00:34:17 +03:00
Aliaksandr Valialkin
b986516fbe
lib/storage: create and use lib/uint64set
instead of map[uint64]struct{}
...
This should improve inverted index search performance for filters matching big number of time series,
since `lib/uint64set.Set` is faster than `map[uint64]struct{}` for both `Add` and `Has` calls.
See the corresponding benchmarks in `lib/uint64set`.
2019-09-24 21:17:55 +03:00
Aliaksandr Valialkin
ef2296e420
lib/storage: typo fix: return dstData instead of data from mergeTagToMetricIDsRows
2019-09-24 19:32:34 +03:00
Aliaksandr Valialkin
a6086cde78
lib/storage: limit the number of metricIDs in tag->metricIDs row
...
This reduces the overhead on index and metaindex in lib/mergeset
2019-09-24 00:49:51 +03:00
Aliaksandr Valialkin
c9063ece66
lib/storage: share tsids across all the partSearch instances
...
This should reduce memory usage when big number of time series matches the given query.
2019-09-23 22:35:15 +03:00
Aliaksandr Valialkin
4e26ad869b
lib/{storage,mergeset}: verify PrepareBlock callback results
...
Do not touch the first and the last item passed to PrepareBlock
in order to preserve sort order of mergeset blocks.
2019-09-23 20:43:13 +03:00
Aliaksandr Valialkin
0772191975
lib/mergeset: detect whether we are in test by executable suffix
2019-09-22 23:12:15 +03:00
Aliaksandr Valialkin
48999e5396
lib/workingsetcache: remove data race when resetting c.misses
2019-09-22 19:36:49 +03:00
Aliaksandr Valialkin
0adebae1f8
lib/storage: generate the first tag->metricIDs item in a mergeset block with a single metricID
...
The first item from each mergeset block goes into index (lib/mergeset.blockHeader),
so it must be short in order to reduce index size.
2019-09-22 19:21:33 +03:00
Aliaksandr Valialkin
0686ac52c3
lib/{storage,mergeset}: merge tag->metricID
rows into tag->metricIDs
rows for common tag
values
...
This should improve lookup performance if the same `label=value` pair exists
in big number of time series.
This should also reduce memory usage for mergeset data cache, since `tag->metricIDs` rows
occupy less space than the original `tag->metricID` rows.
2019-09-20 22:06:41 +03:00
Aliaksandr Valialkin
68722c3c74
lib/encoding: optimize UnmarshalUint* and UnmarshalInt*
2019-09-20 13:08:16 +03:00
Aliaksandr Valialkin
a544f49c2b
lib/storage: optimize selecting all the metricIDs by scanning MetricID->TSID entries instead of tag->MetricID entries
...
The number of MetricID->TSID entries is smaller than the number of tag->MetricID entries
and MetricID->TSID entries are usually shorter than tag->MetricID entries.
This should improve performance when selecting all the metricIDs.
2019-09-20 11:54:10 +03:00
Aliaksandr Valialkin
00cfb2d2b9
lib/mergeset: rename misleading mergeSmallParts to mergeExistingParts
2019-09-19 21:48:20 +03:00
Aliaksandr Valialkin
37dc223e25
lib/mergeset: use sort.IsSorted instead of sort.SliceIsSorted in inmemoryBlock.isSorted in order to reduce memory allocations
2019-09-19 20:13:08 +03:00
Aliaksandr Valialkin
a84fe76677
lib/storage: use sort.Sort instead of sort.slice in getSortedMetricIDs
2019-09-19 20:07:22 +03:00
Aliaksandr Valialkin
3a697a935a
lib/storage: skip duplicate call to intersectMetricIDsWithTagFilter on zero successful intersects
2019-09-19 17:49:56 +03:00
Aliaksandr Valialkin
51a21c7d4b
lib/mergeset: fill partHeader.firstItem on first block flush
2019-09-19 17:48:09 +03:00
Aliaksandr Valialkin
3d83f5d334
lib/storage: mark tag filter returning errFallbackToMetricNameMatch as useless
...
This will save CPU on subsequent calls for this filter
2019-09-18 19:10:32 +03:00
Aliaksandr Valialkin
8d35718dc6
lib/storage: properly construct keys for uselessTagFiltersCache and register useless negative tag filters there
2019-09-17 23:20:27 +03:00
Aliaksandr Valialkin
0c8057924f
lib/mergeset: properly check for sorted block headers
...
Fix a typo for https://github.com/VictoriaMetrics/VictoriaMetrics/issues/181
2019-09-13 21:59:29 +03:00
hanzai
e2274714b1
lib/workingsetcache: adjust switching from mode=split
to mode=whole
smoothly and load cachefile successfully
2019-09-13 19:13:01 +03:00
Aliaksandr Valialkin
bad53e4207
lib/mergeset: dynamically calculate the maximum number of items per part, which can be cached in OS page cache
2019-09-11 14:53:45 +03:00
Aliaksandr Valialkin
88f8670ede
lib/fs: add MustStopDirRemover for waiting until pending directories are removed on graceful shutdown
...
This patch is mainly required for laggy NFS. See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/162
2019-09-05 11:13:17 +03:00
Aliaksandr Valialkin
9eb5de334f
lib/storage: typo fix
2019-09-04 19:58:01 +03:00
Aliaksandr Valialkin
16dd145586
lib/storage: remove duplicate tag keys on MetricName.Marshal
call
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/172
2019-09-04 18:13:45 +03:00
Aliaksandr Valialkin
4862e93024
lib/fs: try harder with directory removal on NFS in the event of temporary lock
...
Do not give up after 11 attempts of directory removal on laggy NFS.
Add `vm_nfs_dir_remove_failed_attempts_total` metric for counting the number of failed attempts
on directory removal.
Log failed attempts on directory removal after long sleep times.
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/162
2019-09-04 12:24:50 +03:00
Aliaksandr Valialkin
e1d76ec1f3
lib/storage: invalidate tagFilters -> TSIDS
cache when newly added index data becomes visible to search
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/163
2019-08-29 15:08:35 +03:00
Aliaksandr Valialkin
aeaa5de5fe
lib/prombp: apply ba06b47c16
...
The following commands used:
gofmt -r '(uint64(x)&0x7F)<<shift -> uint64(x&0x7F)<<shift' -w ./lib/prompb/
gofmt -r '(int64(x)&0x7F)<<shift -> int64(x&0x7F)<<shift' -w ./lib/prompb/
2019-08-29 13:35:27 +03:00
Aliaksandr Valialkin
9196c085a7
all: port to FreeBSD on GOARCH=amd64
2019-08-28 01:19:23 +03:00
Aliaksandr Valialkin
2655220c58
lib/storage: go fmt
2019-08-27 14:29:51 +03:00
Aliaksandr Valialkin
bf915fc0db
lib/storage: report proper maxMetrics limit when more than -search.maxUniqueTimeseries series match the given filters
2019-08-27 14:21:42 +03:00
Aliaksandr Valialkin
2fc157ff7a
lib/storage: properly handle (?i)
in the tag filter regexp
...
Fixes https://github.com/VictoriaMetrics/VictoriaMetrics/issues/161
2019-08-26 00:44:45 +03:00
Aliaksandr Valialkin
0dc0006f34
lib/storage: calculate the maximum number of rows per small part from -memory.allowedPercent
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/159
This simplifies error detection additionally to the `vm_rows_ignored_total` counters.
2019-08-25 15:31:47 +03:00
Aliaksandr Valialkin
4b688fffee
lib/storage: calculate the maximum number of rows per small part from -memory.allowedPercent
...
This should improve query speed over recent data on machines with big amounts of RAM
2019-08-25 14:41:12 +03:00
Aliaksandr Valialkin
1402a6b981
lib/storage: properly limit the number of output rows in small and big parts storage
...
Previously small parts storage didn't take into account the available disk space for big parts.
2019-08-25 14:41:12 +03:00
Aliaksandr Valialkin
3308279c4e
lib/storage: remove outdated comment on maxRowsPerSmallPart
...
The commend became outdated after the commit ed6ac1a5df027f0dfc22448e3b27c26b6f77c67a,
which stops merging of small parts on graceful shutdown instead of waiting
for their completion.
2019-08-25 13:47:32 +03:00
Aliaksandr Valialkin
ac004665b5
all: return 503 http error if service is temporarily unavailable
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/156
2019-08-23 09:55:07 +03:00
Aliaksandr Valialkin
8c03a8c4b4
app/vminsert: allow setting the maximum number of labels per time series via -maxLabelsPerTimeseries
2019-08-23 08:45:26 +03:00
Aliaksandr Valialkin
380cae23a0
lib/storage: add benchmarks for regexp filter match / mismatch
...
These benchmarks allow estimate the performance of regexp filters in promql
2019-08-22 16:36:42 +03:00
Aliaksandr Valialkin
4f738c8a15
lib/storage: try slower path for searching the tag filter with the minimum number of matching time series before giving up with increase -search.maxUniqueTimeseries
error
2019-08-19 16:04:21 +03:00
Aliaksandr Valialkin
c23b66a1ad
lib/storage: pre-allocate memory for blockHeader slice in unmarshalBlockHeaders
...
This reduces memory usage and memory fragmentation when working with big number of time series
2019-08-19 12:46:33 +03:00
Aliaksandr Valialkin
483de1cc06
lib/workingsetcache: automatically detect when it is better to double cache capacity
2019-08-15 22:57:55 +03:00
Aliaksandr Valialkin
5b41122292
lib/storage: properly cache tagFilters -> TSIDs entries from historical index
2019-08-14 02:29:58 +03:00
Aliaksandr Valialkin
964c296f96
lib/storage: compress contents of cache for tagFilters -> TSIDs
...
This should increase cache capacity
2019-08-14 02:29:52 +03:00
Aliaksandr Valialkin
09fc6e22e5
all: use workingsetcache instead of fastcache
...
This should reduce the amount of RAM required for processing time series
with non-zero churn rate.
The previous cache behavior can be restored with `-cache.oldBehavior` command-line flag.
2019-08-13 21:39:34 +03:00
Aliaksandr Valialkin
99c37c2c96
lib/fs: add test for IsTemporaryFileName
2019-08-13 21:33:45 +03:00
Aliaksandr Valialkin
ec1b185991
lib/storage: remove broken BenchmarkIndexDBSearchTSIDs
2019-08-13 20:22:08 +03:00
Aliaksandr Valialkin
0967683ae9
lib: move common code for creating flock.lock file into fs.CreateFlockFile
2019-08-13 01:45:46 +03:00
Aliaksandr Valialkin
5d8d110010
lib/fs: atomically create file with the given contents on WriteFileAtomically
...
This should prevent from `transaction` and `metadata.json` files corruption
on unclean shutdown such as OOM, `kill -9`, power loss, etc.
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/148
2019-08-12 15:02:55 +03:00
Aliaksandr Valialkin
0b488f1e37
lib/storage: do not change timestamps to constant rate if values are constant or have constant delta
...
This breaks the original timestamps, which results in issues like
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/120 and
https://github.com/VictoriaMetrics/VictoriaMetrics/issues/141 .
2019-08-06 15:40:07 +03:00
Aliaksandr Valialkin
b8bb74ffc6
app/vmstorage: add vm_concurrent_addrows_*
metrics for tracking concurrency for Storage.AddRows calls
...
Track also the number of dropped rows due to the exceeded timeout
on concurrency limit for Storage.AddRows. This number is tracked in `vm_concurrent_addrows_dropped_rows_total`
2019-08-06 15:08:33 +03:00
Aliaksandr Valialkin
8822079b77
lib/storage: properly reset partSearch.fetchData
in partSearch.reset
2019-08-05 09:56:06 +03:00
Aliaksandr Valialkin
47e4b50112
app/vmselect: optimize /api/v1/series
by skipping storage data
...
Fetch and process only time series metainfo.
2019-08-04 23:01:28 +03:00
Aliaksandr Valialkin
10caad4728
lib/decimal: modernize tests a bit
2019-07-31 21:10:03 +03:00
Aliaksandr Valialkin
c14fd6c43f
lib/storage: typo fixes after a77e88db7d
2019-07-30 15:38:52 +03:00
Aliaksandr Valialkin
a77e88db7d
lib/storage: fix matching against tag filter with empty name
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/137
2019-07-30 15:15:09 +03:00
Aliaksandr Valialkin
f586e1f83c
lib/storage: add metrics for calculating skipped rows outside the retention
...
The metrics are:
- vm_too_big_timestamp_rows_total
- vm_too_small_timestamp_rows_total
2019-07-26 14:11:01 +03:00
Aliaksandr Valialkin
b339890dca
lib/encoding/zstd: go fmt
2019-07-25 01:37:16 +03:00
Aliaksandr Valialkin
6c4ca89d75
lib/encoding/zstd: disable CRC checks in pure Go
build
...
This should give slightly better compression and decompressions performance.
Additionally this shaves off 4 bytes per each compressed block.
2019-07-24 19:17:16 +03:00
Aliaksandr Valialkin
d07f616609
lib/encoding: small fixes in tests after the PR #114
2019-07-23 19:37:51 +03:00
Roman Khavronenko
5bf4e5ffb5
all: add Pure Go build (pull request #114 )
...
Updates #94
2019-07-23 19:26:39 +03:00
Aliaksandr Valialkin
0f2ecde0e6
lib/encoding: improve gauge series detection
...
- Series with negative values are always gauges
- Counters may only have increasing values with possible counter resets
This should improve compression ratio for gauge series which
were previously mistakenly detected as counters.
2019-07-20 14:05:09 +03:00
Aliaksandr Valialkin
daba0cdb05
lib/netutil: do not count timeouts as network errors
2019-07-15 23:05:35 +03:00
Aliaksandr Valialkin
e75d5f47c4
lib/storage: remove unused function isTooBigTimeRangeForDateMetricIDs
2019-07-12 02:28:23 +03:00
Aliaksandr Valialkin
fc90ebf43c
lib/storage: do not reduce maxMetrics
on time ranges exceeding maxDaysForDateMetricIDs
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/95
2019-07-12 02:20:34 +03:00
Aliaksandr Valialkin
2bd1a01d1a
lib/storage: do not pollute inverted index with data for samples outside the retention period
2019-07-11 17:04:56 +03:00
Aliaksandr Valialkin
d031e04023
lib/storage: use fast path for orSuffix when searching for metricIDs against plain tag value
2019-07-11 14:48:37 +03:00
Aliaksandr Valialkin
43ea4ce428
lib/storage: remember and skip individual tag filters matching too many metrics
...
This saves CPU time by skipping useless matching for individual tag filters.
2019-07-11 14:48:30 +03:00
Aliaksandr Valialkin
1fe6d784d8
all: consistency renaming: bytesSize -> sizeBytes
2019-07-10 00:47:36 +03:00
Aliaksandr Valialkin
56c154f45b
all: add vm_data_size_bytes
metrics for easy monitoring of on-disk data size and on-disk inverted index size
2019-07-04 19:42:30 +03:00
Aliaksandr Valialkin
c7034fc51b
lib/memory: attempt #3 to determine memory limit for LXC container
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/84
2019-07-01 14:01:13 +03:00
Aliaksandr Valialkin
a41955863a
lib/mergeset: make fmt
2019-06-29 14:25:26 +03:00
Aliaksandr Valialkin
2ecb117082
lib/storage: skip non-matching metricIDs in sortedFilter
...
This should improve performance for big sorteFilter lists.
2019-06-29 13:48:32 +03:00
Aliaksandr Valialkin
0c88afa386
lib/mergeset: speed up binarySearchKey by skipping the first item during binary search
2019-06-29 13:45:49 +03:00
Aliaksandr Valialkin
828078eb45
lib/memory: remove TestReadLXCMemoryLimit, since it doesnt work in Travis
2019-06-28 18:22:46 +03:00
Aliaksandr Valialkin
7b59466667
lib/memory: attempt #2 to determine memory limit inside LXC container
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/84
2019-06-28 18:08:15 +03:00
Aliaksandr Valialkin
593bd35aaa
lib/memory: an attempt to read proper memory limit inside LXC container
...
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/84
2019-06-28 15:34:30 +03:00
Aliaksandr Valialkin
c1be1e4342
lib/storage: optimize time series search by regexp filter
...
This should improve search speed on label filters like `{foo=~"bar.+baz"}`
2019-06-27 16:17:43 +03:00
Aliaksandr Valialkin
683bf2a11f
lib/storage: make sure non-nil args are passed to openIndexDB
2019-06-25 20:10:04 +03:00
Aliaksandr Valialkin
eb2283a029
lib/storage: reduce too big maxMetrics in getTagFilterWithMinMetricIDsCountAdaptive
...
This should improve performance on inverted index search for big amount of unique time series
when big -search.maxUniqueTimeseries is set.
2019-06-25 19:55:27 +03:00
Aliaksandr Valialkin
e8377011ab
lib/storage: free up memory from caches owned by indexDB when it is deleted
2019-06-25 14:42:44 +03:00
Aliaksandr Valialkin
33ea2120c3
lib/storage: use unversioned keys for tag cache in extDB
...
Data in ExtDB cannot be changed, so it is OK to use unversioned keys for tag cache.
This should improve performance for index lookups over big amount of time series.
2019-06-25 13:08:58 +03:00
Aliaksandr Valialkin
cf63669303
lib/storage: skip searching in extDB if it doesn't contain items for the given time range
...
This should improve inverted index search performance for big amount
of unique time series when the search is performed only on recent data.
2019-06-25 13:00:37 +03:00
Aliaksandr Valialkin
af2ceaaa0b
lib/storage: mention source parts on merge error
...
This should improve determining broken source part.
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/76
2019-06-24 14:08:43 +03:00
Aliaksandr Valialkin
e88a03323a
all: initial stubs for Windows support; see https://github.com/VictoriaMetrics/VictoriaMetrics/issues/70
2019-06-20 20:07:10 +03:00
Aliaksandr Valialkin
80db24386e
lib/storage: typo fixes found by golangci-lint; updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/69
2019-06-20 14:37:55 +03:00
Aliaksandr Valialkin
296c14317f
lib/netutil: remove unused TCPListener.name; updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/69
2019-06-20 14:36:15 +03:00
Aliaksandr Valialkin
a78b3dba7f
app/vmstorage: add vm_cache_entries{type="storage/hour_metric_ids"}
metric for tracking active time series count
2019-06-19 18:36:47 +03:00
Aliaksandr Valialkin
710d6c33ea
lib/prompb: remove superflouos bytes copying in ReadSnappy
2019-06-18 20:37:51 +03:00
Aliaksandr Valialkin
6d6c9eb1f8
lib/flagutil: remove unused package
2019-06-18 10:43:55 +03:00
Aliaksandr Valialkin
d2c801029b
lib/storage: persist metric ids for the current and the previous hour on graceful shutdown
...
This should improve performance after restart when the db contains a lot of time series
with high time series churn (i.e. metrics from Kubernetes with many pods and frequent deployments)
2019-06-14 07:55:14 +03:00
Aliaksandr Valialkin
26f8d7ea1b
lib/fs: sync parent dir in MustRemoveAll only if it exists
...
The parent directory may be non-existing when the deleted directory
didn't exist before the MustRemoveAll call
2019-06-12 02:14:44 +03:00
Aliaksandr Valialkin
419197ba08
lib/fs: consolidate *RemoveAll* funcs into a single MustRemoveAll func
...
The func syncs parent dir in order to persist directory removal
in the event of power loss
2019-06-12 01:53:46 +03:00