prometheus

Commit Graph

Author	SHA1	Message	Date
Callum Styan	23c382e240	Remote Storage: string interner should not panic in release (#5487 ) * Don't panic if we try to release a string that is not in the interner. * Move seriesMtx locking in QueueManager's StoreSeries function. This stops us from calling release for strings that aren't interned if there's a race between reading a checkpoint and storing new series labels, which could happen during checkpointing or reloading config. Signed-off-by: Callum Styan <callumstyan@gmail.com>	7 years ago
Callum Styan	3639d51eb6	Remote Storage: string interner should not panic in release (#5487 ) * Don't panic if we try to release a string that is not in the interner. * Move seriesMtx locking in QueueManager's StoreSeries function. This stops us from calling release for strings that aren't interned if there's a race between reading a checkpoint and storing new series labels, which could happen during checkpointing or reloading config. Signed-off-by: Callum Styan <callumstyan@gmail.com>	7 years ago
Callum Styan	e87449b59d	Remote Write: Queue Manager specific metrics shouldn't exist if the queue no longer exists (#5445 ) * Unregister remote write queue manager specific metrics when stopping the queue manager. * Use DeleteLabelValues instead of Unregister to remove queue and watcher related metrics when we stop them. Create those metrics in the structs start functions rather than in their constructors because of the ordering of creation, start, and stop in remote storage ApplyConfig. * Add setMetrics function to WAL watcher so we can set the watchers metrics in it's Start function, but not have to call Start in some tests (causes data race). Signed-off-by: Callum Styan <callumstyan@gmail.com>	7 years ago
Vasily Sliouniaev	ef19ede14a	Prevent reshard concurrent with calling stop (#5460 ) * Prevent reshard concurrent with calling stop Signed-off-by: Vasily <v.sliouniaev@gmail.com>	7 years ago
Vasily Sliouniaev	5be9a1426f	Prevent reshard concurrent with calling stop (#5460 ) * Prevent reshard concurrent with calling stop Signed-off-by: Vasily <v.sliouniaev@gmail.com>	7 years ago
Tom Wilkie	807fd33ecc	Review feedback. - Update read path to use labels.Labels. - Fix the tests. - Remove pack. - Remove unused function. - Fix race in tests. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Callum Styan	1a7923dde3	Add ref counting to string interning so we can remove a string when there are no longer any refs. Add tests for interning. Co-authored-by: Tom Wilkie <tom.wilkie@gmail.com> Signed-off-by: Callum Styan <callumstyan@gmail.com>	7 years ago
Tom Wilkie	c7b3535997	Use pkg/relabelling in remote write. - Unmarshall external_labels config as labels.Labels, add tests. - Convert some more uses of model.LabelSet to labels.Labels. - Remove old relabel pkg (fixes #3647). - Validate external label names. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Tom Wilkie	2fa93595d6	More WAL remote_write tweaks. (#5300 ) * Consistently pre-lookup the metrics for a given queue in queue manager. * Don't open the WAL (for writing) in the remote_write code. * Add some more logging. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Tariq Ibrahim	ab8e9b7423	fix typo in queue_manager.go comment (#5294 ) Signed-off-by: tariqibrahim <tariq181290@gmail.com>	7 years ago
Tom Wilkie	67da8e7b46	Refactor and fix queue resharding (#5286 ) - Remove prometheus_remote_queue_last_send_timestamp_seconds metric. Its not particularly useful, we have highest_timestamp_seconds. - Factor out maxGauage, a gauge that only increases. - Change sharding calculations to use max samples in timestamp - max samples out timestamp (not rates). - Also include the ratio of samples dropped to correctly predict number of pending samples. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Callum Styan	b8106dd459	Review feedback: - Add a dropped samples EWMA and use it in calculating desired shards. - Update metric names and a log messages. - Limit number of entries in the dedupe logging middleware to prevent potential OOM. Signed-off-by: Callum Styan <callumstyan@gmail.com> Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Tom Wilkie	f795942572	Decrement pending sample when queue exits. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Tom Wilkie	efbd9559f4	Deal with corruptions in the WAL: - If we're replaying the WAL to get series records, skip that segment when we hit corruptions. - If we're tailing the WAL for samples, fail the watcher. - When the watcher fails, restart from the latest checkpoint - and only send new samples by updating startTime. - Tidy up log lines and error handling, don't return so many errors on quiting. - Expect EOF when processing checkpoints. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Tom Wilkie	d6f911b511	Factor out logging ratelimit & dedupe middleware. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Tom Wilkie	37ad4db485	Export timestamps in seconds since epoch. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
JoeWrightss	362873f72b	Fix .Log() error message (#5257 ) Signed-off-by: zhoulin xie <zhoulin.xie@daocloud.io>	7 years ago
Callum Styan	37e35f9e0c	Various improvements to WAL based remote write. - Use the queue name in WAL watcher logging. - Don't return from watch if the reader error was EOF. - Fix sample timestamp check logic regarding what samples we send. - Refactor so we don't need readToEnd/readSeriesRecords - Fix wal_watcher tests since readToEnd no longer exists Signed-off-by: Callum Styan <callumstyan@gmail.com>	7 years ago
Tom Wilkie	b93bafeee1	Various fixes to locking & shutdown for WAL-based remote write. - Remove datarace in the exported highest scrape timestamp. - Backoff on enqueue should be per-sample - reset the result for each sample. - Remove diffKeys, unused ctx and cancelfunc in WALWatcher, 'name' from writeTo interface, and pass it to constructor. - Reorder functions in WALWatcher depth-first according to call graph. - Fix vendor/modules.txt. - Split out the various timer periods into consts at the top of the file. - Move w.currentSegmentMetric.Set close to where we set the currentSegment. - Combine r.Next() and isClosed(w.quit) into a single loop. - Unnest some ifs in WALWatcher.watch, propagate erros in decodeRecord, add some new lines to make it easier to read. - Reorganise checkpoint handling to reduce nesting and make it easier to follow. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	7 years ago
Callum Styan	6f69e31398	Tail the TSDB WAL for remote_write This change switches the remote_write API to use the TSDB WAL. This should reduce memory usage and prevent sample loss when the remote end point is down. We use the new LiveReader from TSDB to tail WAL segments. Logic for finding the tracking segment is included in this PR. The WAL is tailed once for each remote_write endpoint specified. Reading from the segment is based on a ticker rather than relying on fsnotify write events, which were found to be complicated and unreliable in early prototypes. Enqueuing a sample for sending via remote_write can now block, to provide back pressure. Queues are still required to acheive parallelism and batching. We have updated the queue config based on new defaults for queue capacity and pending samples values - much smaller values are now possible. The remote_write resharding code has been updated to prevent deadlocks, and extra tests have been added for these cases. As part of this change, we attempt to guarantee that samples are not lost; however this initial version doesn't guarantee this across Prometheus restarts or non-retryable errors from the remote end (eg 400s). This changes also includes the following optimisations: - only marshal the proto request once, not once per retry - maintain a single copy of the labels for given series to reduce GC pressure Other minor tweaks: - only reshard if we've also successfully sent recently - add pending samples, latest sent timestamp, WAL events processed metrics Co-authored-by: Chris Marchbanks <csmarchbanks.com> (initial prototype) Co-authored-by: Tom Wilkie <tom.wilkie@gmail.com> (sharding changes) Signed-off-by: Callum Styan <callumstyan@gmail.com>	7 years ago
Simon Pasquier	f678e27eb6	: use latest release of staticcheck (#5057 ) : use latest release of staticcheck It also fixes a couple of things in the code flagged by the additional checks. Signed-off-by: Simon Pasquier <spasquie@redhat.com> Use official release of staticcheck Also run 'go list' before staticcheck to avoid failures when downloading packages. Signed-off-by: Simon Pasquier <spasquie@redhat.com>	7 years ago
Bartek Płotka	62c8337e77	Moved configuration into `relabel` package. (#4955 ) Adapted top dir relabel to use pkg relabel structs. Removal of this in a separate tracked here: https://github.com/prometheus/prometheus/issues/3647 Signed-off-by: Bartek Plotka <bwplotka@gmail.com>	7 years ago
Ryota Arai	135d580ab2	Introduce min_shards for remote write to set minimum number of shards. (#4924 ) Signed-off-by: Ryota Arai <ryota.arai@gmail.com>	7 years ago
Ben Kochie	c6399296dc	Fix spelling/typos (#4921 ) * Fix spelling/typos Fix spelling/typos reported by codespell/misspell. * UK -> US spelling changes. Signed-off-by: Ben Kochie <superq@gmail.com>	7 years ago
fengyuceNv	94fff219ab	improve remote storage enqueue performance (#4772 ) Signed-off-by: fyc <fyc22788@ly.com>	7 years ago
Daisy T	7d01ead689	change time.duration to model.duration for standardization (#4479 ) Signed-off-by: Daisy T <daisyts@gmx.com>	7 years ago
Goutham Veeramachaneni	c28cc5076c	Saner defaults and metrics for remote-write (#4279 ) * Rename queueCapacity to shardCapacity * Saner defaults for remote write * Reduce allocs on retries Signed-off-by: Goutham Veeramachaneni <cs14btech11014@iith.ac.in>	8 years ago
Bryan Boreham	3277aeefaa	Add queue name to logger for remote writes More than one remote_write destination can be configured, in which case it's essential to know which one each log message refers to. Signed-off-by: Bryan Boreham <bjboreham@gmail.com>	8 years ago
Tom Wilkie	b58199bf12	Review feedback. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	8 years ago
Tom Wilkie	3353bbd018	Add proper unclean shutdown handling with a cancellable context. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	8 years ago
Tom Wilkie	e51d6c4b6c	Make remote flush deadline a command line param. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	8 years ago
Tom Wilkie	a6c353613a	Make the flush deadline configurable. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	8 years ago
Tom Wilkie	aa17263edd	Remove WaitGroup and extra goroutine. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	8 years ago
Tom Wilkie	f3c61f8bb2	Only give remote queues 1 minute to flush samples on shutdown. Signed-off-by: Tom Wilkie <tom.wilkie@gmail.com>	8 years ago
Mario Trangoni	464e747f1e	fix some comments typos (#4059 )	8 years ago
Tom Wilkie	fdb574b608	Review feedback.	8 years ago
Tom Wilkie	97a5fc8cbb	Correctly stop the timer used in the remote write path.	8 years ago
Tom Wilkie	22d820ef8e	Review feedback.	8 years ago
Tom Wilkie	f8c9d375b6	Correctly stop the timer used in the remote write path.	8 years ago
ferhat elmas	ffa673f7d8	General simplifications (#3887 ) Another try as in #1516	8 years ago
Bryan Boreham	8a4535e6ad	Re-use timer instead of creating new ones on every sample The docs for `time.After()` note that "The underlying Timer is not recovered by the garbage collector until the timer fires".	8 years ago
Tom Wilkie	6e4d4ea402	Initialise some counters in remote storage API.	8 years ago
Tom Wilkie	ee011d906d	Port remote read server to 2.0.	8 years ago
Tom Wilkie	8fe0212ff7	Port 'Make queue manager configurable.' to 2.0, see #2991	8 years ago
Brian Brazil	73dc96e7f5	Fix leak of ticker in remote storage queue manager.	8 years ago
Tom Wilkie	4f8efdbd59	Prevent number of remote write shards from going negative. This can happen in the situation where the system scales up the number of shards massively (to deal with some backlog), then scales it down again as the number of samples sent during the time period is less than the number received.	8 years ago
Fabian Reinartz	d21f149745	*: migrate to go-kit/log	8 years ago
Tom Wilkie	5169f990f9	Review feedback: add yaml struct tags, don't embed queue config. Also, rename QueueManageConfig to QueueConfig, for consistency with tags.	9 years ago
Tom Wilkie	454b661145	Make queue manager configurable.	9 years ago
Tom Wilkie	ec999ff397	Prevent number of remote write shards from going negative. This can happen in the situation where the system scales up the number of shards massively (to deal with some backlog), then scales it down again as the number of samples sent during the time period is less than the number received.	9 years ago

1 2

75 Commits (be67b8d460cb30b94de771df24753816443993a0)