Commit Graph

423 Commits

Author SHA1 Message Date
beorn7 ee1078a03c Move registry hashing to xxhash
This is a much stronger hash function than fnv64a and comparably fast
(with super-fast assembly implementation for amd64).

Performance is not critical here anyway.

The old fnv64a is kept for vectors, where collision detection is in
place and the weakness of the hashing doesn't matter that much. I
implemented a vector version with xxhash and found that xxhash is
slower in all cases except very very high cardinality (where it is
only slightly faster). Also, ``xxhash.New`` comes with an allocation
of 80 bytes. Thus, to keep vectors alloc-free, we needed to add a
`sync.Pool`, which would have an additional performance overhead.

Signed-off-by: beorn7 <beorn@grafana.com>
2019-10-14 21:18:38 +02:00
beorn7 c2e3855f3b Minimal “fix” for hash collisions
This makes the collisions a bit less likely by XOR'ing descIDs rather
than adding them up for the collectorID.

Signed-off-by: beorn7 <beorn@grafana.com>
2019-10-14 20:14:43 +02:00
beorn7 bf9ff715fe Expose bug #633
Signed-off-by: beorn7 <beorn@grafana.com>
2019-10-14 20:02:58 +02:00
Xin Zhang 7c6a9d7568 Fix check status code
Signed-off-by: Xin Zhang <ahjdzx1990@gmail.com>
2019-10-12 16:08:00 +08:00
beorn7 ad58180fde Add WriteHeader call to Flush
Flush is another of the methods that will call WriteHeader if it
hasn't happened yet. Since we want to call observeWriteHeader (if
set), we need to do the WriteHeader call already here, similar to what
we have done in Write and ReadFrom.

This commit also adds comments explaining the above to not tempt
developers to remove the WriteHeader call.

Signed-off-by: beorn7 <beorn@grafana.com>
2019-08-19 18:19:23 +02:00
Andrey Yurchenkov b0bcec8f2e
Fix typo in documentation
Signed-off-by: Andrey Yurchenkov <painhardcore@gmail.com>
2019-08-13 13:43:39 +03:00
beorn7 4b95c4ab42 Support new base64 encoding for pushing to the Pushgateway
This should only be released after PGW 0.9 is released.

Signed-off-by: beorn7 <beorn@grafana.com>
2019-07-22 20:31:37 +02:00
beorn7 48cd700822 Use method constants from the http package
Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-28 15:23:00 +02:00
beorn7 9a1440d469 Add Delete method
Fixes #611.

Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-28 15:12:09 +02:00
beorn7 2f3a0f8f2e Make the AlreadyRegisteredError useful for wrapped registries
Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-14 17:55:35 +02:00
beorn7 f61dbeaded Update doc comment of NewProcessCollector
- Now also works on MS Windows.
- The hints for updating from older versions is obsolete by now.

Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-14 12:33:52 +02:00
Calle Pettersson c5f4190338 Implement process collector for Windows (#596)
* Implement process collector for Windows

Signed-off-by: Calle Pettersson <calle@cape.nu>
2019-06-14 12:26:55 +02:00
beorn7 e7f6132a76 Remove work duplication in doc comment
Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-12 01:46:17 +02:00
beorn7 761a2ff07c Remove all deprecated features
This is in preparation of the upcoming v1 release.

Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-11 16:28:47 +02:00
beorn7 8576729b6f Add a simple buildInfoCollector
Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-06 22:24:23 +02:00
beorn7 4d8144c11a Add test for the new promhttp_metric_handler_errors_total metric
Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-06 16:56:58 +02:00
beorn7 3fa9fca2cb Add an error counter for internal errors in the HTTP handler
The doc comments explain the rationale in a quite detailed way.

Fixes #543 and #542

Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-05 20:07:49 +02:00
beorn7 3d1d93eb51 Remove use of deprecated procfs functions
Signed-off-by: beorn7 <beorn@grafana.com>
2019-06-04 19:26:20 +02:00
Björn Rabenstein 388f9869ac
Merge pull request #588 from prometheus/beorn7/test
Attempt to fix flakiness of TestGoCollectorMemStats
2019-05-26 23:20:33 +02:00
beorn7 c818d96e15 Unflake TestGoCollectorGC
It is perfectly possible that a normal GC happens just before the
forced one. Thus seeing 2 GCs is fine.

Whenever this test failed, it was because two GCs were seen.

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-26 14:11:51 +02:00
beorn7 fe90eea9bb Attempt to fix flakiness of TestGoCollectorMemStats
This is really lame as it essentially just uses longer times to
wait. The test is still timing-dependent and thus could still
theoretically fail with unlucky scheduling. However, we are testing
something that _is_ about timing. Turning this all into something not
timing-dependent would be first quite involved and second might defeat
the purpose of testing code that is inherently about timing.

Let's see how this works out in practice.

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-26 13:25:59 +02:00
beorn7 b5f69192ee Purge remaining references to v0.10 from doc comments
Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-16 23:35:36 +02:00
beorn7 bc54582c5e Make use of pre-existing context in InstrumentRoundTripperTrace
Fixes #580

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-16 17:52:19 +02:00
beorn7 6c520f6aca Add test to expose #580
Tests are heavily inspired by @shturec, see #584.

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-16 17:52:19 +02:00
Björn Rabenstein 775f94f6fc
Merge pull request #581 from prometheus/beorn7/test
Fix race in TestGoCollectorMemStats
2019-05-16 15:31:37 +02:00
beorn7 e91cd81632 Remove outdated use-case description from doc comment
We stopped advertising binary-wide setting of a label quite a while
ago. This doc comment was missed in the cleanup.

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-16 11:23:48 +02:00
beorn7 709dc4b1de Fix race in TestGoCollectorMemStats
Related to #573 (but only partially fixes it).

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-16 10:37:37 +02:00
Camilo Viecco facfcc21f8 fix regression issue 574
Signed-off-by: Camilo Viecco <camilo_viecco1@symantec.com>
2019-05-14 11:44:55 -07:00
beorn7 886e2ee0c0 Clarify deprecation of `DefObjectives`
Previously, the whole `Objectives` field was marked as deprecated by
linters.

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-08 13:38:43 +02:00
Björn Rabenstein 2d3b0fe0e0
Merge pull request #568 from prometheus/beorn7/go-collector
Return previous memstats if Go collector needs longer than 1s
2019-05-07 00:10:32 +02:00
Bjoern Rabenstein 7cf0955421 Handle long ReadMemStats duration in Go collector
tl;dr: Return previous memstats if reading new ones takes longer than
1s.

See the doc comment of NewGoCollector for details.

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-06 23:28:31 +02:00
beorn7 547c945a62 Replace fmt.Sprintf by simple concatenation
In this simple case, it's the fastest and easiest.

Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-06 23:28:31 +02:00
beorn7 016273b1f9 Fix and tweak Go collector tests
Signed-off-by: beorn7 <bjoern@rabenste.in>
2019-05-06 23:28:31 +02:00
Björn Rabenstein f1d50bcc55
Merge pull request #559 from nghialt/master
Add HTTPClient interface to Pusher struct
2019-05-06 01:31:30 +02:00
NghiaLT 6ea6f07918 Replace HTTPClient with HTTPDoer; Update document accordingly
Signed-off-by: NghiaLT <nghialt.11@gmail.com>
2019-05-06 04:28:14 +07:00
Björn Rabenstein 906d2976cb
Merge pull request #567 from prometheus/beorn7/counter
Make TestCounterAddLarge more robust
2019-05-04 18:08:04 +02:00
Bjoern Rabenstein bf1f4e4a24 Make TestCounterAddLarge more robust
The previous `float64(math.MaxUint64 + 1)` is too close to
`float64(math.MaxUint64)` to actually overflow as indended.

The counter code is actually converting forward and backward and
compare the original and twice-converted value. On most platform, this
will create a deviation and thus trigger the expected behavior. By
sheer "luck", one might end up with the same value and thus still use
the uint64 representation. Which is OK within the precision we can
expect. But it breaks the test. With this change, the next
representable floating point value greater than the floating point
value used to represent math.MaxUint64 is used.

Signed-off-by: Bjoern Rabenstein <bjoern@rabenste.in>
2019-05-03 22:25:32 +02:00
Bjoern Rabenstein 57f7bd35fd Remove remaining traces of pre-go-1.8 handling
Signed-off-by: Bjoern Rabenstein <bjoern@rabenste.in>
2019-05-03 21:30:30 +02:00
KevinBetterQ a325fb7566 fix: fix a typo
Signed-off-by: KevinBetterQ <1093850932@qq.com>
2019-05-01 17:50:23 +08:00
Bjoern Rabenstein 1173d73405 Increase minimum required Go version to 1.9
This allows us to simplify a bunch of code while still supporting the
last four Go minor versions.

We have also run into minor annoyances a couple of times by now to
keep supporting 1.7 and 1.8.

It's time to pull the plug!

Signed-off-by: Bjoern Rabenstein <bjoern@rabenste.in>
2019-04-28 23:28:57 +02:00
NghiaLT c9d77912ac Add HTTPClient interface to Pusher struct
Signed-off-by: NghiaLT <nghialt.11@gmail.com>
2019-04-26 16:53:21 +07:00
Simon Pasquier 6894bb3c7c Add .golangci.yml
Signed-off-by: Simon Pasquier <spasquie@redhat.com>
2019-04-25 10:38:07 +02:00
Simon Pasquier 15da21f1e4 *: replace golang.org/x/net/context by context
The context package is available since Go 1.7 which is the minimal version
supported by client_golang.

Signed-off-by: Simon Pasquier <spasquie@redhat.com>
2019-04-09 15:10:51 +02:00
Sergiusz Urbaniak 60728f445d prometheus/promhttp: actually observe values in trace in example
This is similar to
https://github.com/prometheus/client_golang/pull/535 fixing ExampleInstrumentRoundTripperDuration.

Signed-off-by: Sergiusz Urbaniak <sergiusz.urbaniak@gmail.com>
2019-03-06 13:59:53 +01:00
Frank Spitulski 10e34c668f fix(push): incorporate changes from review
Signed-off-by: Frank Spitulski <fspituls@ucsd.edu>
2019-02-24 15:15:35 -08:00
Frank Spitulski 7b127b3fe5 feat(push): add format builder option
Allow users to specify which format the pusher should push in.

Signed-off-by: Frank Spitulski <fspituls@ucsd.edu>
2019-02-24 12:51:45 -08:00
beorn7 4c99dd6630 Port histogram improvements into noObjectivesSummary
Signed-off-by: beorn7 <beorn@soundcloud.com>
2019-02-11 19:10:17 +01:00
beorn7 295f7e4861 Fix some comment and naming nits as leftover from #536
Signed-off-by: beorn7 <beorn@soundcloud.com>
2019-02-11 18:52:52 +01:00
Pascal S. de Kloe 19ff277483 Save memory on histograms: slightly faster with less code (#536)
Use the highest bit for buffer switch tracking in histograms

Signed-off-by: Pascal S. de Kloe <pascal@quies.net>
2019-02-11 18:29:02 +01:00
kklipsch a11607d7ae fix client trace examples to actually capture metrics
Signed-off-by: kklipsch <kklipsch@example.com>
2019-02-04 20:30:19 +00:00