Commit graph

98 commits

Author SHA1 Message Date
Aliaksandr Valialkin
f5cb213ef9 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 23:59:21 +03:00
Aliaksandr Valialkin
6721e47ae9 app: respect CPU limits set via cgroups
Update GOMAXPROCS to limits set via cgroups. This should reduce CPU trashing and reduce memory usage
for cases when VictoriaMetrics components run in containers with CPU limits.

See https://github.com/VictoriaMetrics/VictoriaMetrics/issues/685
2020-08-11 23:01:03 +03:00
Aliaksandr Valialkin
a455930ab4 app/vmstorage: rename vm_cache_size_entries{type="storage/prefetchedMetricIDs"} to vm_cache_entries{type="storage/prefetchedMetricIDs"} to be consistent with other vm_cache_entries metrics 2020-08-06 16:34:18 +03:00
Aliaksandr Valialkin
a3e91c593b 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 18:27:21 +03:00
Aliaksandr Valialkin
29bbab0ec9 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 20:02:22 +03:00
Aliaksandr Valialkin
b8303afcd8 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 01:40:38 +03:00
Aliaksandr Valialkin
31ef39e8da lib/httpserver: log remote address in error message from httpserver.Errorf
This should improve detection of the root cause of errors.
Thanks to Anant for the idea.
2020-07-20 14:06:29 +03:00
Aliaksandr Valialkin
0bff96fe4b 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 19:44:04 +03:00
Aliaksandr Valialkin
d962568e93 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 23:33:46 +03:00
Aliaksandr Valialkin
2784015a4d all: print --help output to stdout instead of stderr
This is easier to grep and pipe
2020-05-16 12:03:06 +03:00
Aliaksandr Valialkin
1e5c1d7eaa 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 14:12:24 +03:00
Aliaksandr Valialkin
d6b9a49481 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 13:46:57 +03:00
Aliaksandr Valialkin
f7753b1469 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 12:13:32 +03:00
Aliaksandr Valialkin
a53e332a93 app/vmstorage: add missing shutdown for http server on graceful shutdown
This could result in the following panic during graceful shutdown when `/metrics` page is requested:

http: panic serving 10.101.66.5:57366: runtime error: invalid memory address or nil pointer dereference
goroutine 2050 [running]:
net/http.(*conn).serve.func1(0xc00ef22000)
	net/http/server.go:1772 +0x139
panic(0xa0fc00, 0xe91d80)
	runtime/panic.go:973 +0x3e3
github.com/VictoriaMetrics/VictoriaMetrics/lib/workingsetcache.(*Cache).UpdateStats(0x0, 0xc0000516c8)
	github.com/VictoriaMetrics/VictoriaMetrics/lib/workingsetcache/cache.go:224 +0x37
github.com/VictoriaMetrics/VictoriaMetrics/lib/storage.(*indexDB).UpdateMetrics(0xc00b931d00, 0xc02c41acf8)
	github.com/VictoriaMetrics/VictoriaMetrics/lib/storage/index_db.go:258 +0x9f
github.com/VictoriaMetrics/VictoriaMetrics/lib/storage.(*Storage).UpdateMetrics(0xc0000bc7e0, 0xc02c41ac00)
	github.com/VictoriaMetrics/VictoriaMetrics/lib/storage/storage.go:413 +0x4c5
main.registerStorageMetrics.func1(0x0)
	github.com/VictoriaMetrics/VictoriaMetrics/app/vmstorage/main.go:186 +0xd9
main.registerStorageMetrics.func3(0xc00008c380)
	github.com/VictoriaMetrics/VictoriaMetrics/app/vmstorage/main.go:196 +0x26
main.registerStorageMetrics.func7(0xc)
	github.com/VictoriaMetrics/VictoriaMetrics/app/vmstorage/main.go:211 +0x26
github.com/VictoriaMetrics/metrics.(*Gauge).marshalTo(0xc000010148, 0xaa407d, 0x20, 0xb50d60, 0xc005319890)
	github.com/VictoriaMetrics/metrics@v1.11.2/gauge.go:38 +0x3f
github.com/VictoriaMetrics/metrics.(*Set).WritePrometheus(0xc000084300, 0x7fd56809c940, 0xc005319860)
	github.com/VictoriaMetrics/metrics@v1.11.2/set.go:51 +0x1e1
github.com/VictoriaMetrics/metrics.WritePrometheus(0x7fd56809c940, 0xc005319860, 0xa16f01)
	github.com/VictoriaMetrics/metrics@v1.11.2/metrics.go:42 +0x41
github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver.writePrometheusMetrics(0x7fd56809c940, 0xc005319860)
	github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver/metrics.go:16 +0x44
github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver.handlerWrapper(0xb5a120, 0xc005319860, 0xc005018f00, 0xc00002cc90)
	github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver/httpserver.go:154 +0x58d
github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver.gzipHandler.func1(0xb5a120, 0xc005319860, 0xc005018f00)
	github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver/httpserver.go:119 +0x8e
net/http.HandlerFunc.ServeHTTP(0xc00002d110, 0xb5a660, 0xc0044141c0, 0xc005018f00)
	net/http/server.go:2012 +0x44
net/http.serverHandler.ServeHTTP(0xc004414000, 0xb5a660, 0xc0044141c0, 0xc005018f00)
	net/http/server.go:2807 +0xa3
net/http.(*conn).serve(0xc00ef22000, 0xb5bf60, 0xc010532080)
	net/http/server.go:1895 +0x86c
created by net/http.(*Server).Serve
	net/http/server.go:2933 +0x35c
2020-04-02 21:09:55 +03:00
Aliaksandr Valialkin
3b744f3c32 app/vmstorage: typo fix 2020-04-01 23:43:09 +03:00
Aliaksandr Valialkin
f838cdc86e app/vmstorage: add vm_free_disk_space_bytes metric for monitoring the remaining disk space at -storageDataPath 2020-04-01 23:10:44 +03:00
Aliaksandr Valialkin
8939c19281 app/vmstorage: return 500 status code instead of 200 status code on internal errors inside /snapshot/* handlers 2020-03-10 23:54:27 +02:00
Aliaksandr Valialkin
cf9aee4ec3 all: properly split vm_deduplicated_samples_total among cluster components
Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/345
2020-02-27 23:47:51 +02:00
Aliaksandr Valialkin
1010a57882 all: allow setting flags via environment vars
Now flags can be set via environment vars with the same names as flags.
Command-line flags override flags set via env vars.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/311
2020-02-10 13:31:21 +02:00
Aliaksandr Valialkin
ea66212c93 lib/storage: move -dedup.minScrapeInterval flag outside lib/storage, so it doesnt show up in vminsert in cluster version 2020-02-10 13:07:25 +02:00
Aliaksandr Valialkin
4ed5e9a7ce lib/storage: pre-fetch metricNames for the found metricIDs in Search.Init
This should speed up Search.NextMetricBlock loop for big number of found time series.
2020-01-30 15:16:16 +02:00
Aliaksandr Valialkin
ea53a21b02 all: consistently log durations in seconds with millisecond precision
This should improve logs readability
2020-01-22 18:35:24 +02:00
Aliaksandr Valialkin
4e22b521c2 lib/storage: remove metricID with missing metricID->metricName entry
The metricID->metricName entry can be missing in the indexdb after unclean shutdown
when only a part of entries for new time series is written into indexdb.

Recover from such a situation by removing the broken metricID. New metricID
will be automatically created for time series with the given metricName
when new data point will arive to it.
2019-12-02 20:52:13 +02:00
Aliaksandr Valialkin
d297b65089 lib/storage: add vm_cache_size_bytes{type="storage/hour_metric_ids"} metric 2019-11-13 20:26:05 +02:00
Aliaksandr Valialkin
494ad0fdb3 lib/storage: remove inmemory index for recent hour, since it uses too much memory
Production workload shows that the index requires ~4Kb of RAM per active time series.
This is too much for high number of active time series, so let's delete this index.

Now the queries should fall back to the index for the current day instead of the index
for the recent hour. The query performance for the current day index should be good enough
given the 100M rows/sec scan speed per CPU core.
2019-11-13 18:08:58 +02:00
Aliaksandr Valialkin
633dd81bb5 lib/storage: add -disableRecentHourIndex flag for disabling inmemory index for recent hour
This may be useful for saving RAM on high number of time series aka high cardinality
2019-11-13 15:10:12 +02:00
Aliaksandr Valialkin
f1620ba7c0 lib/storage: fix inmemory inverted index issues found in v1.29
Issues fixed:
- Slow startup times. Now the index is loaded from cache during start.
- High memory usage related to superflouos index copies every 10 seconds.
2019-11-13 13:35:38 +02:00
Aliaksandr Valialkin
87b39222be Revert "lib/fs: do not postpone directory removal on NFS error"
This reverts commit 21aeb02b46649ac9906cb37733f7b155a77a0db9.
2019-11-12 16:29:50 +02:00
Aliaksandr Valialkin
c48e39eea9 lib/storage: add tests for dateMetricIDCache 2019-11-11 13:21:05 +02:00
Aliaksandr Valialkin
5f52eb7653 lib/fs: do not postpone directory removal on NFS error
Continue trying to remove NFS directory on temporary errors for up to a minute.

The previous async removal process breaks in the following case during VictoriaMetrics start

- VictoriaMetrics opens index, finds incomplete merge transactions and starts replaying them.
- The transaction instructs removing old directories for parts, which were already merged into bigger part.
- VictoriaMetrics removes these directories, but their removal is delayed due to NFS errors.
- VictoriaMetrics scans partition directory after all the incomplete merge transactions are finished
  and finds directories, which should be removed, but weren't still removed due to NFS errors.
- VictoriaMetrics panics when it finds unexpected empty directory.

Updates https://github.com/VictoriaMetrics/VictoriaMetrics/issues/162
2019-11-10 13:27:16 +02:00
Aliaksandr Valialkin
9ea2bd822e lib/storage: implement per-day inverted index 2019-11-10 00:20:32 +02:00
Aliaksandr Valialkin
dea2f3efed lib/storage: use specialized cache for (date, metricID) entries
This improves ingestion performance.
2019-11-09 23:09:18 +02:00
Aliaksandr Valialkin
46e67bb78c lib/storage: export vm_new_timeseries_created_total metric for determining time series churn rate 2019-11-08 19:58:21 +02:00
Aliaksandr Valialkin
0063c857f5 lib/storage: add inmemory inverted index for the last hour
It should improve performance for `last N hours` dashboards with update intervals smaller than 1 hour.
2019-11-08 19:37:46 +02:00
Aliaksandr Valialkin
1c777e0245 lib/storage: substitute error message about unsorted items in the index block after metricIDs merge with counter
The origin of the error has been detected and documented in the code,
so it is enough to export a counter for such errors at `vm_index_blocks_with_metric_ids_incorrect_order_total`,
so it could be monitored and alerted on high error rates.

Export also the counter for processed index blocks with metricIDs - `vm_index_blocks_with_metric_ids_processed_total`,
so its' rate could be compared to `rate(vm_index_blocks_with_metric_ids_incorrect_order_total)`.
2019-11-06 14:32:41 +02:00
Aliaksandr Valialkin
6ab9c98a1e app/vmstorage: add -bigMergeConcurrency and -smallMergeConcurrency flags for tuning the maximum number of CPU cores used during merges 2019-10-31 16:17:29 +02:00
Aliaksandr Valialkin
2c654258ef 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 11:17:17 +03:00
Aliaksandr Valialkin
f56c1298ad 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 15:08:43 +03:00
Aliaksandr Valialkin
8253790157 app/vmstorage: consistency renaming for ignored rows metrics
vm_too_big_timestamp_rows_total -> vm_rows_ignored_total{reason="big_timestamp"}
  vm_too_small_timestamp_rows_total -> vm_rows_ignored_total{reason="small_timestamp"}
2019-07-26 20:02:24 +03:00
Aliaksandr Valialkin
c6bec48927 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 14:11:56 +03:00
Aliaksandr Valialkin
ba8195c58e all: consistency renaming: bytesSize -> sizeBytes 2019-07-10 00:47:42 +03:00
Aliaksandr Valialkin
41f512af1c all: add vm_data_size_bytes metrics for easy monitoring of on-disk data size and on-disk inverted index size 2019-07-04 19:43:04 +03:00
Aliaksandr Valialkin
a0c22a6830 app/vmstorage: add vm_cache_entries{type="storage/hour_metric_ids"} metric for tracking active time series count 2019-06-19 18:37:38 +03:00
Aliaksandr Valialkin
d54f5fec0b lib/storage: skip adaptive searching for tag filter matching the minimum number of metrics if the identical previous search didn't found such filter
This should improve speed for searching metrics among high number of time series
with high churn rate like in big Kubernetes clusters with frequent deployments.
2019-06-10 14:07:47 +03:00
Aliaksandr Valialkin
4c3913290a app/vmstorage: add missing _total suffixes to newly added metrics 2019-06-09 22:11:41 +03:00
Aliaksandr Valialkin
d882afa905 lib/storage: optimize time series lookup for recent hours when the db contains many millions of time series with high churn rate (aka frequent deployments in Kubernetes) 2019-06-09 19:14:04 +03:00
Aliaksandr Valialkin
24578b4bb1 all: open-sourcing cluster version 2019-05-23 00:25:38 +03:00
Aliaksandr Valialkin
1836c415e6 all: open-sourcing single-node version 2019-05-23 00:18:06 +03:00