When the exemplar type is an int, it incorrectly gets converted to a 0
when DoubleValue() is called on the exemplar. This adds a check to
ensure that the value is converted properly based on the type.
Signed-off-by: Charlie Le <charlie_le@apple.com>
* docs: 2 to 3 migration guide
Signed-off-by: Jan Fajerski <jfajersk@redhat.com>
* docs/stability: add 3.0 section
Signed-off-by: Jan Fajerski <jfajersk@redhat.com>
* docs/migration: details on enabling legacy name validation
Signed-off-by: Owen Williams <owen.williams@grafana.com>\
* migration: add log format and `le` normalization
Signed-off-by: Jan Fajerski <jfajersk@redhat.com>
* migration: add new enable_http2 default for remote write
Signed-off-by: Jan Fajerski <jfajersk@redhat.com>
---------
Signed-off-by: Jan Fajerski <jfajersk@redhat.com>
Signed-off-by: Owen Williams <owen.williams@grafana.com>
Co-authored-by: Owen Williams <owen.williams@grafana.com>
I used these wrapper methods during initial development of the custom
handler that the deduper now implements. Since the deduper implements
slog.Handler and can be used directly as a logger, these wrapper methods
are no longer needed.
Signed-off-by: TJ Hoplock <t.hoplock@gmail.com>
This change should have been included in the initial prometheus slog
conversion, but I must've lost track of it in all the rebases involved
in that PR.
This changes the dedupe logger so that the only method that needs to use
the lock is the `Handle()` method that actually interacts with the
deduplication map.
Ex:
```
==================
WARNING: DATA RACE
Write at 0x00c000518bc0 by goroutine 29481:
github.com/prometheus/prometheus/util/logging.(*Deduper).WithAttrs()
/home/tjhop/go/src/github.com/prometheus/prometheus/util/logging/dedupe.go:89 +0xef
log/slog.(*Logger).With()
/home/tjhop/.asdf/installs/golang/1.23.1/go/src/log/slog/logger.go:132 +0x106
github.com/prometheus/prometheus/storage/remote.NewQueueManager()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/queue_manager.go:483 +0x7a9
github.com/prometheus/prometheus/storage/remote.(*WriteStorage).ApplyConfig()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/write.go:201 +0x102c
github.com/prometheus/prometheus/storage/remote.(*Storage).ApplyConfig()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/storage.go:92 +0xfd
github.com/prometheus/prometheus/storage/remote.TestWriteStorageApplyConfigsDuringCommit.func1()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/storage_test.go:172 +0x3e4
github.com/prometheus/prometheus/storage/remote.TestWriteStorageApplyConfigsDuringCommit.gowrap1()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/storage_test.go:174 +0x41
Previous read at 0x00c000518bc0 by goroutine 31261:
github.com/prometheus/prometheus/util/logging.(*Deduper).Handle()
/home/tjhop/go/src/github.com/prometheus/prometheus/util/logging/dedupe.go:82 +0x2b1
log/slog.(*Logger).log()
/home/tjhop/.asdf/installs/golang/1.23.1/go/src/log/slog/logger.go:257 +0x228
log/slog.(*Logger).Error()
/home/tjhop/.asdf/installs/golang/1.23.1/go/src/log/slog/logger.go:230 +0x3d4
github.com/prometheus/prometheus/tsdb/wlog.(*Watcher).loop()
/home/tjhop/go/src/github.com/prometheus/prometheus/tsdb/wlog/watcher.go:254 +0x2db
github.com/prometheus/prometheus/tsdb/wlog.(*Watcher).Start.gowrap1()
/home/tjhop/go/src/github.com/prometheus/prometheus/tsdb/wlog/watcher.go:227 +0x33
Goroutine 29481 (running) created at:
github.com/prometheus/prometheus/storage/remote.TestWriteStorageApplyConfigsDuringCommit()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/storage_test.go:164 +0xe4
testing.tRunner()
/home/tjhop/.asdf/installs/golang/1.23.1/go/src/testing/testing.go:1690 +0x226
testing.(*T).Run.gowrap1()
/home/tjhop/.asdf/installs/golang/1.23.1/go/src/testing/testing.go:1743 +0x44
Goroutine 31261 (running) created at:
github.com/prometheus/prometheus/tsdb/wlog.(*Watcher).Start()
/home/tjhop/go/src/github.com/prometheus/prometheus/tsdb/wlog/watcher.go:227 +0x177
github.com/prometheus/prometheus/storage/remote.(*QueueManager).Start()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/queue_manager.go:934 +0x304
github.com/prometheus/prometheus/storage/remote.(*WriteStorage).ApplyConfig()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/write.go:232 +0x151b
github.com/prometheus/prometheus/storage/remote.(*Storage).ApplyConfig()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/storage.go:92 +0xfd
github.com/prometheus/prometheus/storage/remote.TestWriteStorageApplyConfigsDuringCommit.func1()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/storage_test.go:172 +0x3e4
github.com/prometheus/prometheus/storage/remote.TestWriteStorageApplyConfigsDuringCommit.gowrap1()
/home/tjhop/go/src/github.com/prometheus/prometheus/storage/remote/storage_test.go:174 +0x41
==================
--- FAIL: TestWriteStorageApplyConfigsDuringCommit (2.26s)
testing.go:1399: race detected during execution of test
FAIL
FAIL github.com/prometheus/prometheus/storage/remote 68.321s
```
Signed-off-by: TJ Hoplock <t.hoplock@gmail.com>
Remote-write creates several shards to parallelise sending, each with
its own http connection. We do not want them all combined onto one
socket by http2.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
Call through to the underlaying parser if we are not in a histogram
and the entry is a series or exponential native histogram. Otherwise store
and retrieve CT for NHCB.
* fix(omparser): losing exemplars when CT is parsed
Fixes: #15137
Ignore exemplars while peeking ahead during CT parsing.
Simplify state reset with defer().
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
Fix some edge cases when OOO is enabled
Signed-off-by: Vanshikav123 <vanshikav928@gmail.com>
Signed-off-by: Vanshika <102902652+Vanshikav123@users.noreply.github.com>
Signed-off-by: Jesus Vazquez <jesusvzpg@gmail.com>
Co-authored-by: Jesus Vazquez <jesusvzpg@gmail.com>
* NHCB: scrape use state field and not booleans
From comment https://github.com/prometheus/prometheus/pull/14978#discussion_r1800898724
Also make compareLabels read only and move storeLabels to the first
processed classic histogram series.
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
* Speed up TestConvertClassicHistogramsToNHCB 3x
Reduce the startup time and timeouts
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
* lint fix
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
---------
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
* promtool: Add debug flag for rule tests
This makes it print out the tsdb state (both input_series and rules that
are run) at the end of a test, making reasoning about tests much easier.
Signed-off-by: David Leadbeater <dgl@dgl.cx>
* Reuse generated test name from junit testing
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
---------
Signed-off-by: David Leadbeater <dgl@dgl.cx>
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
Co-authored-by: David Leadbeater <dgl@dgl.cx>
turn some loops into subtests to make use of t.Parallel()
requires Go 1.22 to make use of https://go.dev/blog/loopvar-preview
Signed-off-by: machine424 <ayoubmrini424@gmail.com>
scrape: Remove implicit fallback to the Prometheus text format
Remove implicit fallback to the Prometheus text format in case of invalid/missing Content-Type and fail the scrape instead. Add ability to specify a `fallback_scrape_protocol` in the scrape config.
---------
Signed-off-by: alexgreenbank <alex.greenbank@grafana.com>
Signed-off-by: Alex Greenbank <alex.greenbank@grafana.com>
Co-authored-by: Björn Rabenstein <beorn@grafana.com>
When handling recoded histogram chunks the min time of the chunk is
updated by mistake. It should only update when the chunk is completely new.
Otherwise the ongoing chunk's meta will be later than the previously
written samples in it.
Same bug as https://github.com/prometheus/prometheus/pull/14629
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
When handling recoded histogram chunks the min time of the chunk is
updated by mistake. It should only update when the chunk is completely
new.
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
Ayoub Mrini