Kiu/VictoriaMetrics
Kiu/VictoriaMetrics
1
0
Fork 0
mirror of https://github.com/VictoriaMetrics/VictoriaMetrics.git synced 2024-12-30 07:40:06 +01:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
4,980 Commits 113 Branches 748 Tags 745 MiB
4d27fa41c8
Commit Graph

538 Commits

Author SHA1 Message Date
Aliaksandr Valialkin
d7be2753c0 lib/storage: substitute GetTSDBStatusForDate with GetTSDBStatusWithFiltersForDate with nil tfss 2021-05-13 09:02:33 +03:00
Aliaksandr Valialkin
a22a17dc66 lib/storage: merge getTSDBStatusForDate with getTSDBStatusWithFiltersForDate 
These functions are non-trivial, while their code has minimal differences.
It is better from maintainability PoV to merge these functions into a single function.
 2021-05-12 17:56:53 +03:00
Aliaksandr Valialkin
832651c6c2 app/vmselect: follow up after 8a0678678b 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1168
 2021-05-12 17:18:30 +03:00
Nikolay
8a0678678b
Adds tsdb match filters (#1282) 
* init work on filters

* init propose for status filters

* fixes tsdb status
adds test

* fix bug

* removes checks from test
 2021-05-12 15:18:45 +03:00
Aliaksandr Valialkin
33f7bacb01 lib/storage: properly apply time range when matching an empty filter 
It must match all the time series on the given time range.
Previously it was matched to all the time series without the restriction on the given time range.
 2021-05-11 01:12:05 +03:00
Aliaksandr Valialkin
75c2c813fc lib/storage: remove dead code after the commit 3ccf7ea20c 2021-05-08 20:14:11 +03:00
Aliaksandr Valialkin
12d733dd5d app/vminsert: add support for data ingestion via other vminsert nodes 2021-05-08 19:52:57 +03:00
Aliaksandr Valialkin
76027093ca lib/storage: use WARNING instead of INFO level for logging dropped labels 2021-05-03 13:56:43 +03:00
Nikolay
477369b62f
adds stalePartsRemover (#1261) 
for new created partitions
 2021-05-03 11:34:00 +03:00
Aliaksandr Valialkin
58d1e6eeea lib/storage: log dropped labels if the number of labels in a metric exceeds -maxLabelsPerTimeseries command-line flag value 
This should improve debuggability for this case.
 2021-05-01 09:27:55 +03:00
Aliaksandr Valialkin
87179c6839 lib/{storage,mergeset}: fix unaligned 64-bit atomic operation panic for 32-bit architectures 
The panic has been introduced in 56b6b893ce
 2021-04-27 16:41:32 +03:00
Aliaksandr Valialkin
56b6b893ce lib/mergeset: split rows ingestion among multiple shards 
This improves rows ingestion on systems with many CPU cores by reducing lock contention.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1244

Thanks to @waldoweng for the original idea and draft implementation at https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1243
 2021-04-27 15:36:34 +03:00
Aliaksandr Valialkin
f89c1f7f49 lib/storage: typo fix in info message when deleting the part outside the configured retention 
Previously the message was displaying incorrect retention time
 2021-04-27 13:32:46 +03:00
Aliaksandr Valialkin
bbebdf9ba1 lib/{storage,mergeset}: remove empty directories on startup. Such directories can be left after unclean shutdown on NFS storage 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1142
 2021-04-22 13:02:44 +03:00
Aliaksandr Valialkin
251747e253 lib/storage: code clarification: remove caching the found metricName in searchMetricName 2021-04-13 10:22:21 +03:00
Aliaksandr Valialkin
df32d2836c lib/storage: properly handle big time ranges passed to /api/v1/labels and /api/v1/label/<labelName>/values 
It should be faster querying all the labels and/or all the values instead of querying per-day labels/values on time ranges exceeding maxDaysForPerDaySearch
 2021-04-07 13:33:46 +03:00
Aliaksandr Valialkin
dc9eafcd02 app/{vminsert,vmagent}: add -sortLabels command-line option for sorting time series labels before ingesting them in the storage 
This option can be useful when samples for the same time series are ingested with distinct order of labels.
For example, metric{k1="v1",k2="v2"} and metric{k2="v2",k1="v1"}.
 2021-03-31 23:27:58 +03:00
Aliaksandr Valialkin
e1f699bb6c lib/storage: reduce memory usage when ingesting samples for the same time series with distinct order of labels 2021-03-31 21:24:46 +03:00
Aliaksandr Valialkin
2601cc0fb0 lib/storage: do not update b.nextIdx if no samples are removed because of retention 2021-03-29 12:00:21 +03:00
Aliaksandr Valialkin
aa81039b42 app/vmselect: log the metric which trigger rollup result cache reset 
This should help finding the source of stale metrics
 2021-03-25 21:31:39 +02:00
Aliaksandr Valialkin
6e855d4b82 lib/storage: tune loopsCountPerMetricNameMatch according to production workload 2021-03-25 13:27:47 +02:00
Aliaksandr Valialkin
3a3d2165f9 lib/storage: do not reload metricName for the same metricID in Search.NextMetricBlock 
This should speed up Search.NextMetricBlock a bit
 2021-03-23 17:56:49 +02:00
Aliaksandr Valialkin
3cfb3a3683 lib/storage: respect the deadline passed to Storage.SearchMetricNames 2021-03-22 23:03:17 +02:00
Aliaksandr Valialkin
8e2afdf568 lib/storage: improve Search.NextMetricBlock performance by using MetricID->MetricName cache 2021-03-22 22:49:18 +02:00
Aliaksandr Valialkin
910092ca4d lib/storage: tune loopsCountPerMetricNameMatch 2021-03-22 12:53:17 +02:00
Aliaksandr Valialkin
726f6ad804 lib/storage: small code simplification after 6cee5338b2 2021-03-18 15:21:13 +02:00
Aliaksandr Valialkin
6cee5338b2 lib/storage: prevent from infinite loop if {__graphite__="..."} filter matches a metric name with *, [ or { chars 
The idea has been borrowed from https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1137
 2021-03-18 14:53:47 +02:00
Aliaksandr Valialkin
45dabfac1b lib/storage: faster move heavy filters to the end of list 2021-03-17 15:12:13 +02:00
Aliaksandr Valialkin
ccfb0ae2d3 lib/storage: limit loops count in order to reduce max CPU usage during filter search 2021-03-17 00:49:26 +02:00
Aliaksandr Valialkin
576a80b3d9 lib/storage: do not modify filterLoopsCount stats with loopsCount stats 
Such a modification can result in incorrect filter sorting later
 2021-03-17 00:49:26 +02:00
Aliaksandr Valialkin
fd86a7dc1d lib/storage: time series search optimization according to production workload profiling 
Do not pass filter metric ids to getMetricIDsForTagFilter, since it has been appeared that this slows down
the function by multiple times when it finds big number of metricIDs (tens of millions).
 2021-03-16 20:01:43 +02:00
Aliaksandr Valialkin
e36fbfae5b lib/storage: further tuning for time series search 2021-03-16 18:46:22 +02:00
Aliaksandr Valialkin
dd7e82c34f app/vmstorage: add -logNewSeries command-line flag for determining the source of series churn rate 2021-03-15 22:38:50 +02:00
Aliaksandr Valialkin
9c77e34ef9 lib/storage: further tuning for time series selector code 2021-03-15 20:31:34 +02:00
Aliaksandr Valialkin
3ccf7ea20c lib/storage: tune per-day index search 2021-03-15 13:31:55 +02:00
Aliaksandr Valialkin
f669531506 lib/storage: further tune filters sorting logic 2021-03-12 00:53:04 +02:00
John Belmonte
364fdf4a56
spelling fix: adjacent (#1115) 2021-03-09 09:18:19 +02:00
Aliaksandr Valialkin
345980f78f lib/storage: go fmt 2021-03-08 12:03:31 +02:00
Aliaksandr Valialkin
18fe0ff14b lib/storage: tune loopsCount estimations in getMetricIDsForTagFilterSlow 
The adjusted estmations give up to 2x lower median response times on 200qps /api/v1/query_range workload
 2021-03-07 21:12:35 +02:00
Aliaksandr Valialkin
4de4da1e2a lib/storage: typo fix: umarshal -> unmarshal 2021-03-02 20:47:59 +02:00
Aliaksandr Valialkin
2c44178645 lib/storage: consistency renaming: durationsPerDateTagFilterCache -> loopsPerDateTagFilterCache 2021-02-23 15:47:19 +02:00
faceair
15d61c4879
lib/storage: correct tagfilter match cost (#1079) 2021-02-22 21:46:56 +02:00
Aliaksandr Valialkin
636c55b526 lib/mergeset: reduce memory usage for inmemoryBlock by using more compact items representation 
This also should reduce CPU time spent by GC, since inmemoryBlock.items don't have pointers now,
so GC doesn't need visiting them.
 2021-02-21 22:06:47 +02:00
Aliaksandr Valialkin
388cdb1980 lib/storage: do not re-calculate stats for heavy tag filters 
This should reduce the number of slow queries when stats for heavy tag filters was recalculated.
 2021-02-21 21:39:01 +02:00
Aliaksandr Valialkin
48656dcc38 lib/{mergeset,storage}: allow merging smaller number of small parts 
While this may increase CPU and disk IO usage needed for background merge,
this also recudes CPU usage during queries in production. This is because
such queries tend to read recently added data and it is better to have lower number
of parts for such data in order to reduce CPU usage.

This partially reverts ebf8da3730
 2021-02-21 21:28:36 +02:00
Aliaksandr Valialkin
cb311bb156 lib/{mergeset,storage}: do not use pools for indexBlock and inmemoryBlock during their caching, since this results in higher memory usage in production without any performance gains 2021-02-21 21:18:59 +02:00
Aliaksandr Valialkin
f9084611bd lib/storage: use composite index for a query with a name filter and negative filters 2021-02-18 18:57:23 +02:00
Aliaksandr Valialkin
a537c4f602 lib/storage: properly handle queries containing a filter on metric name plus any number of negative filters and zero non-negative filters 
Example: `node_cpu_seconds_total{mode!="idle"}`
 2021-02-18 18:46:36 +02:00
Aliaksandr Valialkin
e540c02014 lib/storage: prevent from running identical heavy tag filters in concurrent queries when measuring the number of loops for such tag filter. 
This should reduce CPU usage spikes when measuring the number of loops needed for heavy tag filters
 2021-02-18 13:58:18 +02:00
Aliaksandr Valialkin
711f8a5b8d lib/storage: sort tag filters by the number of loops they need for the execution 
This metric should work better than the filter execution duration, since it cannot be distorted
by concurrently running queries.
 2021-02-18 12:47:38 +02:00
Aliaksandr Valialkin
ce99b48a9a Revert "lib/mergeset: tune lifetime for entries inside block caches" 
This reverts commit 458c89324d.

Production testing revealed zero improvements for memory usage with reduced lifetime for entries in block caches.
 2021-02-17 20:42:21 +02:00
Aliaksandr Valialkin
939d5ffc2b lib/storage: move composite filters to the top during sorting 2021-02-17 20:26:51 +02:00
Aliaksandr Valialkin
faad6f84a4 lib/storage: return back filter arg to getMetricIDsForTagFilter function 
The filter arg has been removed in the commit c7ee2fabb8
because it was preventing from caching the number of matching time series per each tf.

Now the cache contains duration for tf execution, so the filter shouldn't break such caching.
 2021-02-17 19:33:22 +02:00
Aliaksandr Valialkin
d4849561ef app/vmstorage: export vm_composite_filter_success_conversions_total and vm_composite_filter_missing_conversions_total metrics 2021-02-17 19:13:38 +02:00
Aliaksandr Valialkin
33806264ec lib/storage: revert ecf132933e, since negative filters require the same amount of work as non-negative filters 2021-02-17 18:55:04 +02:00
Aliaksandr Valialkin
63fc140624 lib/storage: tag filters sorting... 2021-02-17 17:55:29 +02:00
Aliaksandr Valialkin
74424b55ee lib/storage: further tune tag filters sorting 2021-02-17 17:28:15 +02:00
Aliaksandr Valialkin
442fcfec5a lib/storage: tune the logic for sorting tag filters according the their exeuction times 2021-02-17 15:00:08 +02:00
Aliaksandr Valialkin
4a07820048 lib/storage: make sure that nobody uses partitions when closing the table 2021-02-17 14:59:04 +02:00
Aliaksandr Valialkin
d61f7b7279 lib/storage: more tuning for tag filters sorting according the time they take 2021-02-16 21:22:23 +02:00
Aliaksandr Valialkin
458c89324d lib/mergeset: tune lifetime for entries inside block caches 
This should reduce memory usage in general case without significant CPU usage increase
 2021-02-16 18:11:51 +02:00
Aliaksandr Valialkin
ca191696fe lib/storage: tune sorting for tag filters 2021-02-16 13:04:49 +02:00
Aliaksandr Valialkin
ecf132933e lib/storage: increase match cost for negative tag filters, since they need to scan all the label pairs 2021-02-15 16:34:23 +02:00
Aliaksandr Valialkin
4e39bf148c vendor: update github.com/VictoriaMetrics/metrics from v1.13.1 to v1.14.0 
The new version switches from log-linear histograms to log-based histograms,
which provide up to 3.6 times better accuracy.
 2021-02-15 15:12:29 +02:00
Aliaksandr Valialkin
9f5ac603a7 lib/storage: reduce the minimum supported retention for inverted index from one month to one day 2021-02-15 15:12:29 +02:00
Aliaksandr Valialkin
71c417427c lib/storage: sort tag filters by actual execution time instead of by the number of matching time series 
This should improve query speed for queries with regexp filters matching small number of time series
on a label with big number of unique values.
 2021-02-15 00:18:13 +02:00
Aliaksandr Valialkin
c727d2219b lib/storage: properly hanle regexp tag filters with dots, which can be converted to full string match filters. 
For example `{label=~"foo\.bar"}` should be converted to `{label="foo.bar"}`. Previously it has was mistakenly conveted to `{label="foo\.bar"}` .
This could result in missing time series for such tag filters.
 2021-02-14 23:38:14 +02:00
Aliaksandr Valialkin
0e26b7168a lib/storage: return back in-order applying of tag filters, since concurrently executing tag filters can result in CPU and RAM waste in common case 2021-02-10 22:41:04 +02:00
Aliaksandr Valialkin
b51c23dc5b lib/storage: parallelize tag filters execution a bit 
This should reduce execution time when a query contains multiple tag filters and each such filter matches big number of time series.
 2021-02-10 18:12:25 +02:00
Aliaksandr Valialkin
c7ee2fabb8 lib/storage: remove filter arg from getMetricIDsForDateTagFilter function 
The `filter` arg breaks the logic for sorting tag filters by the matching metrics,
which may result in non-optimal performance during time series search.
 2021-02-10 18:12:20 +02:00
Aliaksandr Valialkin
57cac289e0 lib/storage: fix inconsistencies in error logs 2021-02-10 18:12:16 +02:00
Aliaksandr Valialkin
5d5f0b0627 lib/storage: load metadata before loading indexdb, since indexdb depends on the metadata 2021-02-10 17:55:40 +02:00
Aliaksandr Valialkin
cdecf83ce5 app/vmstorage: export vm_composite_index_min_timestamp metric 2021-02-10 17:14:08 +02:00
Aliaksandr Valialkin
553016ea99 lib/storage: disable composite index usage when querying old data 2021-02-10 14:57:50 +02:00
Aliaksandr Valialkin
fcb7655d1e lib/storage: fix metric name match for composite filter 2021-02-10 01:27:45 +02:00
Aliaksandr Valialkin
c7dccebaef lib/storage: optimize search by label filters matching big number of time series 2021-02-10 00:44:54 +02:00
Aliaksandr Valialkin
6b4e6c229c lib/storage: reduce lock contention in dateMetricIDCache when registering new time series for the current day 
This should help systems with multiple CPU cores
 2021-02-10 00:01:13 +02:00
Aliaksandr Valialkin
d56390b925 optimize Storage.updatePerDateData() 2021-02-09 02:55:36 +02:00
Aliaksandr Valialkin
fda61e8e96 lib/storage: skip deduplication when creating inmemory data blocks 
The deduplication will be performed later during merging such blocks.
 2021-02-09 02:25:32 +02:00
Aliaksandr Valialkin
a4140de9e6 lib/mergeset: unconditionally cache indexdb blocks 
Production workloads show that indexdb blocks must be cached unconditionally for reducing CPU usage.
This shouldn't increase memory usage too much, since unused blocks are removed from the cache every two minutes.
 2021-02-09 00:47:50 +02:00
Aliaksandr Valialkin
cb96a1865b app/vmstorage: export missing vm_cache_size_bytes metrics for indexdb and data caches 2021-02-09 00:47:00 +02:00
Aliaksandr Valialkin
2242647a04 lib/storage: optimize data ingestion in the beginning of every hour 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1046
 2021-02-08 12:01:12 +02:00
Aliaksandr Valialkin
83d3e582ab lib/storage: check for prevHourMetricIDs cache before falling back to checking for (date, metricID) entries during data ingestion 
This should reduce possible CPU usage spikes at the beginning of every hour.

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/1046
 2021-02-04 18:48:13 +02:00
Aliaksandr Valialkin
d16f22f3a1 app/vmselect,lib/storage: properly parse Graphite selectors with inner wildcards 
Example: foo{bar{x,yz},a[b-c],*de}
 2021-02-03 20:14:22 +02:00
Aliaksandr Valialkin
a5a1b9bd66 lib/storage: fix a bug, which breaks searching by Graphite wildcard filters 2021-02-03 20:14:22 +02:00
Aliaksandr Valialkin
6123aa3e75 sort orSuffixes in tagFilter.InitFromGraphiteQuery for faster seeks 2021-02-03 20:14:22 +02:00
Aliaksandr Valialkin
157c02622b app/vmselect: add ability to set Graphite-compatible filter via {__graphite__="foo.*.bar"} syntax 2021-02-03 01:21:54 +02:00
Aliaksandr Valialkin
31ec79eaf6 lib/storage: inline marshalTags function and remove the code for handling duplicate tags from here 
This is a follow-up commit after c8ea697db8
 2021-01-12 15:13:30 +02:00
Aliaksandr Valialkin
c8ea697db8 lib/storage: de-duplicate tags in MetricName.sortTags 
Leave only the last tag among tags with duplicate keys. This is needed for reliable addition of extra_labels
during data ingestion. See https://github.com/VictoriaMetrics/VictoriaMetrics/pull/1007 for details.
 2021-01-12 15:03:42 +02:00
Aliaksandr Valialkin
9dcb18e03d app/vmstorage: disable final merge by default, since it may result in high disk IO and CPU usage without measurable benefits such as increased query performance and reduced disk space usage 2021-01-08 00:16:05 +02:00
Aliaksandr Valialkin
490c69c64e lib/storage: wait for pending transactions before closing and dropping the partition 
This deflakes `make test-full-386` test
 2020-12-25 11:45:53 +02:00
Aliaksandr Valialkin
cab7e936a3 lib/storage: physically remove stale parts 
Previously they were removed from partition struct, but the corresponding directories weren't removed.

This is a follow-up for 46dba00756
 2020-12-24 16:51:36 +02:00
Aliaksandr Valialkin
9e4ed5e591 lib/storage: do not remove parts outside the configured retention if they are currently merged 
These parts are automatically removed after the merge is complete.
 2020-12-24 08:51:28 +02:00
Aliaksandr Valialkin
46dba00756 lib/storage: remove stale parts as soon as they go outside the configured retention 
Previously such parts could remain undeleted for long durations until they are merged with other parts.
This should help for `-retentionPeriod` values smaller than one month.
 2020-12-22 19:54:31 +02:00
Aliaksandr Valialkin
d65c03c004 lib/storage: properly determine max rows for output part when merging small parts 2020-12-18 23:14:38 +02:00
Aliaksandr Valialkin
ebf8da3730 lib/{storage,mergeset}: tune background merge process in order to reduce CPU usage and disk IO usage 2020-12-18 20:01:08 +02:00
Aliaksandr Valialkin
4146fc4668 all: properly handle CPU limits set on the host system/container 
This can reduce memory usage on systems with enabled CPU limits.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/946
 2020-12-08 21:07:29 +02:00
Aliaksandr Valialkin
357f886f97 Revert "lib/storage: add missing (AccountID, ProjectID) in MetricName.String() test" 
This reverts commit f3e80eb70e493620e42a0cc22a62c9af75076c77, since it isn't needed for single-node version of VictoriaMetrics
 2020-11-29 01:27:25 +02:00
Aliaksandr Valialkin
ace969d595 lib/storage: add missing (AccountID, ProjectID) in MetricName.String() test 2020-11-29 01:26:04 +02:00
Aliaksandr Valialkin
8a057e705a lib/storage: log metric name plus all its labels when the metric timestamp is outside the configured retention 
This should simplify debugging when the source of the metric with unexpected timestamp must be found.
 2020-11-25 14:41:37 +02:00
Aliaksandr Valialkin
b65236530c lib/storage: typo fix in error message: allowd->allowed 2020-11-25 14:15:42 +02:00
Aliaksandr Valialkin
78d2715d04 all: spelling fix: superflouos->superfluous. This is a follow-up for 0acdab3ab9 2020-11-24 12:42:22 +02:00
Aliaksandr Valialkin
f765985947 lib/fs: replace fs.OpenReaderAt with fs.MustOpenReaderAt 
All the callers for fs.OpenReaderAt expect that the file will be opened.
So it is better to log fatal error inside fs.MustOpenReaderAt instead of leaving this to the caller.
 2020-11-23 09:57:21 +02:00
Aliaksandr Valialkin
266788be14 app/vmselect: use storage.NewSearchQuery() instead of constructing storage.SearchQuery in-place 
This should prevent from bugs when AccountID and ProjectID aren't set in storage.SearchQuery.
 2020-11-16 18:24:00 +02:00
Aliaksandr Valialkin
0208d8c103 lib/storage: add a test for Storage.SearchMetricNames 2020-11-16 13:15:16 +02:00
Aliaksandr Valialkin
465923b181 app/vmselect/graphite: add /tags/findSeries handler from Graphite Tags API 
See https://graphite.readthedocs.io/en/stable/tags.html#exploring-tags
 2020-11-16 12:53:13 +02:00
Aliaksandr Valialkin
48d033a198 app/vminsert: add /tags/tagSeries and /tags/tagMultiSeries handlers from Graphite Tags API 
See https://graphite.readthedocs.io/en/stable/tags.html#adding-series-to-the-tagdb
 2020-11-16 02:39:58 +02:00
Aliaksandr Valialkin
4aaee33860 lib/storage: do not show artifically created label for reverse Graphite labels at /api/v1/labels page 2020-11-16 00:44:35 +02:00
immerrr again
51c529a2b6
app/vmstorage: add "/internal/force_flush" endpoint (#893) 2020-11-11 14:40:27 +02:00
Aliaksandr Valialkin
368b69b4c4 app/vmselect: properly handle errors in GetLabelsOnTimeRange and GetLabelValuesOnTimeRange 2020-11-05 01:38:38 +02:00
Aliaksandr Valialkin
1cb78ba1a0 lib/storage: remove data race when updating rowsDeleted 2020-11-05 01:12:21 +02:00
Aliaksandr Valialkin
b378cd6ed8 app/vmselect: optimize querying for /api/v1/labels and /api/v1/label/<name>/values when start and end args are set 2020-11-05 01:01:33 +02:00
Aliaksandr Valialkin
caeb74f068 app/vmselect: reduce memory usage when query touches big number of time series 2020-11-04 17:04:04 +02:00
Aliaksandr Valialkin
ae91a6883c lib/{storage,mergeset}: clean cached index blocks and inmemory blocks more aggressively 
Previously such blocks were cleaned after they weren't accessed during 10 minutes.
Now they are cleaned after one minute of missing access. This should reduce memory usage in general case.
 2020-11-04 17:04:04 +02:00
Aliaksandr Valialkin
f3f62ab04e lib/storage: do not report about the need of free disk space if parts cannot be merged due to too big write amplification 2020-11-03 15:32:02 +02:00
Aliaksandr Valialkin
c87fb9191e lib/storage: remove unneeded fmt.Sprintf 2020-11-03 14:20:31 +02:00
Aliaksandr Valialkin
fe289331dd lib/storage: remove obsolete code 2020-11-02 19:11:59 +02:00
Aliaksandr Valialkin
29e4e7f422 lib/storage: drop more samples outside the given retention during background merge 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/17
 2020-11-01 23:35:06 +02:00
Aliaksandr Valialkin
d12c4914f0 lib/storage: properly handle the case when key="__name__" is passed to MetricName.AddTag* 2020-11-01 23:35:06 +02:00
Aliaksandr Valialkin
64e2d66014 lib/storage: code cleanup after 5bfd4e6218 2020-11-01 23:35:06 +02:00
Aliaksandr Valialkin
5bfd4e6218 app/vmstorage: support for -retentionPeriod smaller than one month 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/173
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/17
 2020-10-20 14:31:44 +03:00
Aliaksandr Valialkin
01987f8c77 lib/storage: small code adjustements after d2960a20e0 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/781
 2020-10-17 01:16:54 +03:00
faceair
d2960a20e0
evaluate the execution cost of all tag filters (#824) 
* evaluate the execution cost of all tag filters

* fix suffixes typo
 2020-10-17 00:46:55 +03:00
Aliaksandr Valialkin
68f0e00761 app/vmstorage: add vm_rows_added_to_storage_total metric, which shows the total number of rows added to storage since app start 2020-10-09 13:35:48 +03:00
Aliaksandr Valialkin
af90b3121c app/vmstorage: add -finalMergeDelay command-line flag for configuring the delay before final merge for per-month partitions after no new data is ingested to it 2020-10-07 17:35:44 +03:00
Aliaksandr Valialkin
764dc2499f lib/storage: code cleanup after 10f2eedee0 
Remove the code that uses metricIDs caches for the current and the previous hour during metricIDs search,
since this code became unused after implementing per-day inverted index almost a year ago.

While at it, fix a bug, which could prevent from finding time series with names containing dots (aka Graphite-like names
such as `foo.bar.baz`).
 2020-10-01 19:06:23 +03:00
Aliaksandr Valialkin
10f2eedee0 lib/storage: imrpove cache effectiveness for time series ids matching the given filters 
Previously the maximum cache lifetime has been limited by 10 seconds. Now it is extended up to a day.
This should reduce CPU usage in the following cases:

* when querying recently added data with small churn rate for time series
* when querying historical data
 2020-10-01 14:38:25 +03:00
Aliaksandr Valialkin
bec9b31b81 lib/storage: allow set values higher than 1 for vm_merge_need_free_disk_space if there are multiple partitions with deferred merges due to disk space shortage 2020-09-29 22:51:43 +03:00
Aliaksandr Valialkin
a9db81c4ab app/vmstorage: add metrics for determining whether background merges need additional disk space to complete 
These metrics are:

* vm_small_merge_need_free_disk_space
* vm_big_merge_need_free_disk_space

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/686
 2020-09-29 21:48:33 +03:00
Aliaksandr Valialkin
1659135752 lib/storage: fix tests for 32-bit arches such as GOARCH=386 and GOARCH=arm 2020-09-29 13:10:22 +03:00
Aliaksandr Valialkin
38789e4aa0 lib/storage: fix 32-bit builds for GOARH=386 or GOARCH=arm 2020-09-29 12:40:35 +03:00
Aliaksandr Valialkin
95688cbfc5 all: add native format for data export and import 
The data can be exported via [/api/v1/export/native](https://victoriametrics.github.io/#how-to-export-data-in-native-format) handler
and imported via [/api/v1/import/native](https://victoriametrics.github.io/#how-to-import-data-in-native-format) handler.
 2020-09-27 19:54:07 +03:00
Aliaksandr Valialkin
82973f8ae7 Revert "lib/storage: remove unused fetchData arg from BlockRef.MustReadBlock" 
This reverts commit bab6a15ae0.

Reason for revert: the `fetchData` arg is used in cluster branch.
Leaving this arg in master branch makes smaller the diff with cluster branch.
 2020-09-24 22:44:23 +03:00
Aliaksandr Valialkin
bab6a15ae0 lib/storage: remove unused fetchData arg from BlockRef.MustReadBlock 
This arg became unused after 23bdc1f107
 2020-09-24 20:48:40 +03:00
Aliaksandr Valialkin
24ca30bf66 lib/storage: correctly use maxBlockSize in various checks 
Previously `maxBlockSize` has been multiplied by 8 in certain checks. This is unnecessary.
 2020-09-24 18:12:56 +03:00
Aliaksandr Valialkin
a69234ed18 lib/storage: code prettifying after be5e1222f3 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/781
 2020-09-22 00:36:45 +03:00
faceair
be5e1222f3
add filter to getMetricIDs (#783) 
* add getMetricIDs filter

* check nil filter before use
 2020-09-22 00:33:43 +03:00
Aliaksandr Valialkin
9739283dad lib/storage: reduce CPU load for idle VictoriaMetrics by reducing the frequency for the need for background merges 2020-09-21 15:54:11 +03:00
Aliaksandr Valialkin
26115891db lib/decimal: properly store Inf values 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/752
 2020-09-18 19:07:07 +03:00
Aliaksandr Valialkin
1f33dd717f lib/storage: add /internal/force_merge handler for running forced compactions on historical per-month partitions 
This may be useful for freeing up storage space after time series deletion.

See https://victoriametrics.github.io/#force-merge for more details.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/686
 2020-09-17 12:20:40 +03:00
Aliaksandr Valialkin
8beb0da6ad lib/{mergeset,storage}: compare errors with errors.Is() 2020-09-17 03:03:02 +03:00
Aliaksandr Valialkin
067d7c1ea1 lib/{mergeset,storage}: code prettifying 2020-09-17 02:06:31 +03:00
Aliaksandr Valialkin
020bd8685e lib/storage: removed duplicate checks for empty parts during merge - another check is in the beginning of mergeParts functions 2020-09-17 01:49:03 +03:00
Aliaksandr Valialkin
5a90a92378 lib/storage: do not store inf values, since they may lead to significant precision loss for previously stored values 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/752
 2020-09-11 14:44:53 +03:00
Aliaksandr Valialkin
f6bc608e86 app/vmselect: initial implementation of Graphite Metrics API 
See https://graphite-api.readthedocs.io/en/latest/api.html#the-metrics-api
 2020-09-11 00:30:01 +03:00
Aliaksandr Valialkin
9d8fdff6c5 lib/storage: reuse timestamp blocks for adjancent metric blocks with identical timestamps 
This should reduce disk space usage when scraping targets containing metrics with identical names
such as `node_cpu_seconds_total`, histograms, quantiles, etc.

Expose `vm_timestamps_blocks_merged_total` and `vm_timestamps_bytes_saved_total` metrics for monitoring
the effectiveness of timestamp blocks merging.
 2020-09-09 23:59:32 +03:00
Aliaksandr Valialkin
8d9eb5f808 lib/storage: mention time range used in the query that led to error message 
This should improve detecting slow queries with too big time ranges
 2020-08-10 13:46:36 +03:00
Aliaksandr Valialkin
582c74cd93 lib/storage: mention tag filters used in the query that led to error message 
This should improve detecting invalid or heavy queries that lead to errors.
 2020-08-10 13:36:49 +03:00
Aliaksandr Valialkin
f3d33e23c9 app/vmstorage: improve error logging when the request times out 2020-08-10 13:23:26 +03:00
Aliaksandr Valialkin
84fd8af6d3 lib/storage: slow down concurrent searches when the number of concurrent inserts reaches the limit 
This should improve data ingestion performance when heavy searches are executed

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/648
See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/618
 2020-08-07 08:49:40 +03:00
Aliaksandr Valialkin
9043a509a3 lib/storage: properly check timeouts and pace limits 
Previously they were checked on every iteration for small number of iterations
 2020-08-07 08:40:37 +03:00
Aliaksandr Valialkin
bc8381613d app/vmselect: reduce memory allocations by pre-allocatin memory for time series map and for a list of time series names 2020-08-06 19:17:58 +03:00
Aliaksandr Valialkin
8e44fba76d lib/storage: reduce the frequency (and overhead) for timeout and pace limiter checks by 4x 2020-08-06 18:45:55 +03:00
Aliaksandr Valialkin
ad730d8a17 lib/storage: optimize prefetching metric names for the given metricIDs 2020-08-06 16:53:10 +03:00
Aliaksandr Valialkin
8f16388428 lib/storage: limit the number of concurrent calls to storage.searchTSIDs to GOMAXPROCS*2 
This should limit the maximum memory usage and reduce CPU trashing on vmstorage
when multiple heavy queries are executed.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/648
 2020-08-05 18:30:07 +03:00
Aliaksandr Valialkin
5e71fab8a6 lib/storage: reduce the maximum number of concurrent merge workers to GOMAXPROCS/2 
Previously the limit has been raised to GOMAXPROCS, but it has been appeared that this
increases query latencies since more CPUs are busy with merges.

While at it, substitute `*MergeConcurrencyLimitCh` channels with simple integer limits.
 2020-07-31 17:46:56 +03:00
Aliaksandr Valialkin
e7959094f6 lib/storage: remove prioritizing of merging small parts over merging big parts, since it doesn't work as expected 
The prioritizing could lead to big merge starvation, which could end up in too big number of parts that must be merged into big parts.

Multiple big merges may be initiated after the migration from v1.39.0 or v1.39.1. It is OK - these merges should be finished soon,
which should return CPU and disk IO usage to normal levels.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/648
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/618
 2020-07-30 19:57:27 +03:00
Aliaksandr Valialkin
922d9aadf2 lib/storage: properly update vm_slow_row_inserts_total metric when importing multiple data points per time series at once 
Previously the `vm_slow_row_inserts_total` metric may be incremented multiple times for different data points per a single time series,
while only a single increment is needed when inserting the first data point for this time series.
 2020-07-30 16:17:24 +03:00
Sasasu
5b81bdde39
lib/storage: metaindexRow use memroy more efficiently (#655) 
due to memory align the metaindexRow structure use 64-byte pre object.
this commit changes the order of field, make metaindexRow use 56-byte pre
object.

Signed-off-by: Sasasu <su@sasasu.me>
 2020-07-27 19:02:53 +03:00
Aliaksandr Valialkin
54ef2d8112 lib/storage: slightly reduce code difference between single-node and cluster versions 2020-07-24 00:31:16 +03:00
Aliaksandr Valialkin
039c9d2441 lib/storage: respect -search.maxQueryDuration when searching for time series in inverted index 
Previously the time spent on inverted index search could exceed the configured `-search.maxQueryDuration`.
This commit stops searching in inverted index on query timeout.
 2020-07-23 21:21:42 +03:00
Aliaksandr Valialkin
2a45871823 lib/storage: add more fine-grained pace limiting for search 2020-07-23 19:26:08 +03:00
Aliaksandr Valialkin
6f05c4d351 lib/storage: improve prioritizing of data ingestion over querying 
Prioritize also small merges over big merges.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/291
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/648
 2020-07-23 13:23:36 +03:00
Aliaksandr Valialkin
61c611f5ad lib/storage: properly calculate global metrics in UpdateStats() 2020-07-23 00:35:15 +03:00
Aliaksandr Valialkin
228d137936 lib/storage: reorder mergeBlockStreams() args in order to make them more consistent 2020-07-22 21:58:10 +03:00
Aliaksandr Valialkin
e4303d3d21 lib/storage: prevent possible race condition when all the goroutines exit Storage.AddRows, before goroutines other goroutines are blocked on searchTSIDsCond inside Storage.searchTSIDs 
This condition may occur after the following sequence of events:

1) A goroutine enters the loop body when len(addRowsConcurrencyCh) == cap(addRowsConcurrencyCh) inside Storage.searchTSIDs.
2) All the goroutines return from Storage.AddRows.
3) The goroutine from step 1 blocks on searchTSIDsCond.Wait() inside the loop body.

The goroutine remains blocked until the next call to Storage.AddRows, which calls searchTSIDsCond.Signal().
This may take indefinite time.
 2020-07-22 21:52:34 +03:00
Aliaksandr Valialkin
d3442b40b2 lib/uint64set: optimize adding items to the set via Set.AddMulti 2020-07-21 20:56:59 +03:00
Aliaksandr Valialkin
e1107fec10 lib/storage: reset MetricName->TSID cache after marking metricIDs as deleted 
This is a follow-up commit after 12b16077c4 ,
which didn't reset the `tsidCache` in all the required places.
This could result in indefinite errors like:

    missing metricName by metricID ...; this could be the case after unclean shutdown; deleting the metricID, so it could be re-created next time

Fix this by resetting the cache inside deleteMetricIDs function.
 2020-07-14 14:06:32 +03:00
Aliaksandr Valialkin
cb92113632 lib/storage: limit the maximum concurrency for data ingestion to GOMAXPROCS 
Previously the concurrency has been limited to GOMAXPROCS*2. This had little sense,
since every call to Storage.AddRows is bound to CPU, so the maximum ingestion bandwidth
is achieved when the number of concurrent calls to Storage.AddRows is limited to the number of CPUs,
i.e. to GOMAXPROCS.
 2020-07-08 17:32:18 +03:00
Aliaksandr Valialkin
32b9fb58b8 lib/storage: clarify out of retention period error message by mentioning -retentionPeriod command-line flag 2020-07-08 13:54:26 +03:00
Aliaksandr Valialkin
12b16077c4 lib/storage: reset MetricName->TSID cache after deleting time series 
This should prevent from adding new data points to deleted time series
without the need to check for the deleted time series.

This improves ingestion performance a bit when the `deleted time series ids` aka `dmis` set
contains big number of time series.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/596

Based on the idea from @n4mine at https://github.com/VictoriaMetrics/VictoriaMetrics/pull/604
 2020-07-06 22:01:08 +03:00
Aliaksandr Valialkin
6daa5f7500 lib/storage: prioritize data ingestion over heavy queries 
Heavy queries could result in the lack of CPU resources for processing the current data ingestion stream.
Prevent this by delaying queries' execution until free resources are available for data ingestion.

Expose `vm_search_delays_total` metric, which may be used in for alerting when there is no enough CPU resources
for data ingestion and/or for executing heavy queries.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/291
 2020-07-05 19:42:05 +03:00
Aliaksandr Valialkin
0b2086b7a5 app/vminsert: prevent from adding and/or selecting labels with empty values 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/600
 2020-07-02 23:14:11 +03:00
Aliaksandr Valialkin
d5dddb0953 all: use %w instead of %s for wrapping errors in fmt.Errorf 
This will simplify examining the returned errors such as httpserver.ErrorWithStatusCode .
See https://blog.golang.org/go1.13-errors for details.
 2020-06-30 23:05:11 +03:00
Aliaksandr Valialkin
7532dbcdf5 app/vmselect/promql: properly override label values from group_left and group_right lists like Prometheus does 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/577
 2020-06-21 16:33:01 +03:00
Tristan Su
ac3700ed1e
lib/storage: set big/small merge concurrency (#568) 
fixed #567

Co-authored-by: Tristan Su <suqing.sq@alibaba-inc.com>
 2020-06-19 01:25:48 +03:00
Aliaksandr Valialkin
b542e50680 app/vminsert: export metrics for determining ingested rows with dropped or truncated labels 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/565
 2020-06-19 01:10:37 +03:00
Aliaksandr Valialkin
08495360b0 lib/storage: add key!=".+" filter additionally to negative filter matching empty value such as key!~"|foo" 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/546
 2020-06-18 20:03:48 +03:00
Aliaksandr Valialkin
ae1cc0fc4b lib/storage: properly match {tag!="|foo"} filters 
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/546
 2020-06-10 19:35:56 +03:00
Aliaksandr Valialkin
3d4008263f lib/fs: optimize MustGetFreeSpace performance by caching the results for up to 2 seconds 2020-06-04 13:15:47 +03:00
Aliaksandr Valialkin
a7797dae09 lib/storage: fix Graphite wildcard matching, which has been broken in v1.36.0 
v1.36.0 always returns empty responses for Graphite wildcards like the following

   {__name__=~"foo\\.[^.]*\\.bar\\.baz"}

Temporary workaround for v1.36.0 is to add `[^.]*` to the end of the regexp.
 2020-05-28 12:03:49 +03:00
Aliaksandr Valialkin
d186472081 lib/storage: improve search speed for time series matching Graphite whildcards such as foo.*.bar.baz 
Add index for reverse Graphite-like metric names with dots. Use this index during search for filters
like `__name__=~"foo\\.[^.]*\\.bar\\.baz"` which end with non-empty suffix with dots, i.e. `.bar.baz` in this case.

This change may "hide" historical time series during queries. The workaround is to add `[.]*` to the end of regexp label filter,
i.e. "foo\\.[^.]*\\.bar\\.baz" should be substituted with "foo\\.[^.]*\\.bar\\.baz[.]*".
 2020-05-27 21:45:52 +03:00
Aliaksandr Valialkin
b19ca3eb5f lib/storage: do not increment vm_slow_metric_name_loads_total counter for metric_ids which shouldnt be prefetched, since this may mislead users 2020-05-16 10:21:17 +03:00
Aliaksandr Valialkin
82ffbcb9a6 app/vmstorage: add vm_slow_metric_name_loads_total metric, which could be used as an indicator when more RAM is needed for improving query performance 2020-05-15 14:11:45 +03:00
Aliaksandr Valialkin
82ccdfaa91 app/vmstorage: add vm_slow_row_inserts_total and vm_slow_per_day_index_inserts_total metrics for determining whether VictoriaMetrics required more RAM for the current number of active time series 2020-05-15 13:44:32 +03:00
Aliaksandr Valialkin
0eacea1de1 lib/{storage,mergeset}: further tuning of compression levels depending on block size 
This should improve performance for querying newly added data, since it can be unpacked faster.
 2020-05-15 13:24:37 +03:00
Aliaksandr Valialkin
737d641920 lib/storage: wait for all the goroutines to finish in TestSearch in order to prevent racy behavior on test finish 2020-05-15 13:24:37 +03:00
Aliaksandr Valialkin
4fc33163c4 lib/storage: optimize ingestion pefrormance for new time series 2020-05-15 13:24:37 +03:00
Aliaksandr Valialkin
8b32e7c3a0 lib/storage: reduce indentation in Storage.add 2020-05-15 13:24:37 +03:00
Aliaksandr Valialkin
1573ececb2 lib/storage: return the first error instead of the last error, since the first error usually points to the root cause 2020-05-15 13:24:37 +03:00
Aliaksandr Valialkin
0afd48d2ee lib: extract common code for returning fast unix timestamp into lib/fasttime 2020-05-14 23:02:07 +03:00
Aliaksandr Valialkin
42866fa754 lib/{storage,mergeset}: return dst on error from unmarshalBlockHeaders, so it could be reused 2020-05-14 15:32:07 +03:00
Aliaksandr Valialkin
827a3a7866 lib/storage: document that getnerateUniqueMetricID should return dense ids 2020-05-14 14:08:45 +03:00
Aliaksandr Valialkin
606585f7be lib/{storage,mergeset}: cleanup: remove unused partSearch.indexBlockReuse 2020-05-14 14:03:03 +03:00
Aliaksandr Valialkin
4fe67504f9 lib/storage: optimize label matching for regexp ending with literal suffix 
For example, `{label=~"foo.*bar.+baz"}` contains literal suffix `baz`,
so it should work faster now.
 2020-05-13 11:47:07 +03:00
Aliaksandr Valialkin
3232605524 lib/storage: properly initialize part struct before trying to close it on error 
This should prevent from nil pointer dereference bug at https://github.com/VictoriaMetrics/VictoriaMetrics/issues/468 .
 2020-05-12 14:54:31 +03:00
Aliaksandr Valialkin
dbd0c552d5 lib/storage: gradually pre-populate per-day inverted index for the next day 
This should prevent from CPU usage spikes at 00:00 UTC every day when
inverted index for new day must be quickly created for all the active time series.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/430
 2020-05-12 12:13:05 +03:00
Aliaksandr Valialkin
cc00a2c453 lib/storage: typo fixes in error messages: or -> of 2020-05-12 12:12:42 +03:00
Aliaksandr Valialkin
ce2107bc52 lib/storage: speed up matching for common regexps in label filters 
The following regexps have been optimized:

* 'foo.+bar'
* 'foo.+bar.+baz'

This should improve performance for matching Graphite-like metrics.
 2020-05-11 22:40:55 +03:00
Aliaksandr Valialkin
12a1a71cc1 lib/storage: add a benchmark for Graphite-like regexps for metric names 2020-05-11 22:37:32 +03:00