Commit graph

68 commits

Author SHA1 Message Date
beorn7 9e7c3e3bcd Add the histogram_quantile function.
Since we are now getting really deep into floating point calculation,
the tests had to take into account the precision loss. Since the rule
tests are based on direct line matching in the output, implementing
the "almost equal" semantics was pretty cumbersome, but here we are.
2015-02-22 01:04:51 +01:00
Julius Volz 42601acfde Replace labelsToKey() with metric Fingerprint (fixes grouping bug). 2015-02-21 17:45:47 +01:00
Julius Volz 645cf57bed Fix aggregation grouping key calculation. 2015-02-21 14:05:50 +01:00
Julius Volz 15b2b5aa66 Add tests for invalid uses of "offset". 2015-02-18 02:56:40 +01:00
Julius Volz 72d7b325a1 Implement offset operator.
This allows changing the time offset for individual instant and range
vectors in a query.

For example, this returns the value of `foo` 5 minutes in the past
relative to the current query evaluation time:

    foo offset 5m

Note that the `offset` modifier always needs to follow the selector
immediately. I.e. the following would be correct:

    sum(foo offset 5m) // GOOD.

While the following would be *incorrect*:

    sum(foo) offset 5m // INVALID.

The same works for range vectors. This returns the 5-minutes-rate that
`foo` had a week ago:

    rate(foo[5m] offset 1w)

This change touches the following components:

* Lexer/parser: additions to correctly parse the new `offset`/`OFFSET`
  keyword.
* AST: vector and matrix nodes now have an additional `offset` field.
  This is used during their evaluation to adjust query and result times
  appropriately.
* Query analyzer: now works on separate sets of ranges and instants per
  offset. Isolating different offsets from each other completely in this
  way keeps the preloading code relatively simple.

No storage engine changes were needed by this change.

The rules tests have been changed to not probe the internal
implementation details of the query analyzer anymore (how many instants
and ranges have been preloaded). This would also become too cumbersome
to test with the new model, and measuring the result of the query should
be sufficient.

This fixes https://github.com/prometheus/prometheus/issues/529
This fixed https://github.com/prometheus/promdash/issues/201
2015-02-18 02:41:27 +01:00
Brian Brazil 60271d58bf Change the 2nd argument of round to toNearest.
This is more useful if you want get a multiple of 2 or 5, while
still working for .001.
2015-02-05 16:13:40 +00:00
Marko Mikulicic 8fdacbdf17 Add floor, ceil and round functions. Closes #402 2015-02-04 17:20:56 +01:00
Brian Brazil a31730e88b Make 2nd arg to delta optional. Add a deriv() function.
The 2nd isCounter argument to delta is ugly, make it optional as the first step
of deprecating it. This will makes delta only ever applied to gauges.

Add a deriv function to calculate the least squares
slope of a gauge. This is more useful for prediction than delta,
as it isn't as heavily influenced by outliers at the boundaries.
2015-01-23 14:50:27 +00:00
Bjoern Rabenstein 5859b74f1b Clean up license issues.
- Move CONTRIBUTORS.md to the more common AUTHORS.
- Added the required NOTICE file.
- Changed "Prometheus Team" to "The Prometheus Authors".
- Reverted the erroneous changes to the Apache License.
2015-01-21 20:07:45 +01:00
Julius Volz cc27fb8aab Rename remaining all-caps constants in AST layer.
Change-Id: Ibe97e30981969056ffcdb89e63c1468ea1ffa140
2014-12-25 01:30:47 +01:00
Julius Volz 00a2a93a05 Add regression tests for metrics mutations in AST.
It turned out in the end, that only drop_common_metrics() produced any
erroneous output in the old system. The second expression in the test
("sum(testmetric) keeping_extra") already worked in the old code, but
why not keep it in...

The way to test ranged evaluations is a bit clumsy so far, so I want to
build a nicer test framework in the end, where all the test cases can be
specified as text files which specify desired inputs, outputs, query
step widths, etc.

Change-Id: I821859789e69b8232bededf670a1b76e9e8c8ca4
2014-12-12 20:34:55 +01:00
Bjoern Rabenstein 0ae1d8889a Fix tests after merge.
Change-Id: Ia90da9a3e48ed780ec38c4a6a1fd9ea34e7f6a58
2014-11-25 17:13:04 +01:00
Julius Volz b7bf11230a Add absent() function.
A common problem in Prometheus alerting is to detect when no timeseries
exist for a given metric name and label combination. Unfortunately,
Prometheus alert expressions need to be of vector type, and
"count(nonexistent_metric)" results in an empty vector, yielding no
output vector elements to base an alert on. The newly introduced
absent() function solves this issue:

  ALERT FooAbsent IF absent(foo{job="myjob"}) [...]

absent() has the following behavior:

- if the vector passed to it has any elements, it returns an empty
  vector.

- if the vector passed to it has no elements, it returns a 1-element
  vector with the value 1.

In the second case, absent() tries to be smart about deriving labels of
the 1-element output vector from the input vector:

  absent(nonexistent{job="myjob"})                   => {job="myjob"}
  absent(nonexistent{job="myjob",instance=~".*"})    => {job="myjob"}
  absent(sum(nonexistent{job="myjob"}))              => {}

That is, if the passed vector is a literal vector selector, it takes all
"=" label matchers as the basis for the output labels, but ignores all
non-equals or regex matchers. Also, if the passed vector results from a
non-selector expression, no labels can be derived.

Change-Id: I948505a1488d50265ab5692a3286bd7c8c70cd78
2014-11-25 17:13:04 +01:00
Julius Volz 3d47f94149 Drop metric names after transformations.
After many transformations, it doesn't make sense to keep the metric
names, since the result of the transformation is no longer that metric.
This drops the metric name after such transformations and makes the web
UI deal well with missing metric names.

This depends on the current branch on the following things:

- prometheus/client_golang needs to be at
  e237cf15c6
  in branch "julius/int-fingerprints" (to be merged with new storage)

- prometheus/promdash needs to be at
  dd7691c9c2

Change-Id: Ib3c8cad8d647d9854e8c653c424b8c235ccc231d
2014-11-25 17:13:04 +01:00
Julius Volz 0712d738d1 Allow alternative "by"-clause position in grammar.
In addition to the existing by-clause syntax:

  sum(<expression>) by (<labels>) [keeping_extra]

...this allows the following new syntax:

  sum by (<labels>) [keeping_extra] (<expression>)

Both orderings may be used in a single expression. It is up to the users
to establish guidelines around their usage.

Change-Id: Iba10c9cc5fb6ac62edfcf246d281473e82467992
2014-11-25 17:09:04 +01:00
Julius Volz 0e48c18bbf Allow omitting the metric name in queries.
This allows the following expression syntaxes for selecting timeseries:

  foo                    (already valid before)
  foo{}                  (already valid before)
  {job="prometheus"}     (new, select all timeseries for job "prometheus")

Omitting both the metric name *and* any label matchers ("" or "{}") will
still yield a syntax error.

To get all timeseries, you could do:

    {__name__=~".*"}
       or, without relying on knowledge about __metric__:
    {job=~".*"}

Change-Id: Ifee000b9ac0184ef6ced18411069c7f2699a2dda
2014-11-25 17:09:04 +01:00
Julius Volz 7f5d3c2c29 Fix and improve the fp locker.
Benchmark:
$ go test -bench 'Fingerprint' -test.run 'Fingerprint' -test.cpu=1,2,4

OLD
BenchmarkFingerprintLockerParallel        500000              3618 ns/op
BenchmarkFingerprintLockerParallel-2      100000             12257 ns/op
BenchmarkFingerprintLockerParallel-4      500000             10164 ns/op
BenchmarkFingerprintLockerSerial        10000000               283 ns/op
BenchmarkFingerprintLockerSerial-2      10000000               284 ns/op
BenchmarkFingerprintLockerSerial-4      10000000               288 ns/op

NEW
BenchmarkFingerprintLockerParallel       1000000              1018 ns/op
BenchmarkFingerprintLockerParallel-2     1000000              1164 ns/op
BenchmarkFingerprintLockerParallel-4     2000000               910 ns/op
BenchmarkFingerprintLockerSerial        50000000                56.0 ns/op
BenchmarkFingerprintLockerSerial-2      50000000                47.9 ns/op
BenchmarkFingerprintLockerSerial-4      50000000                54.5 ns/op

Change-Id: I3c65a43822840e7e64c3c3cfe759e1de51272581
2014-11-25 17:07:45 +01:00
Julius Volz e7ed39c9a6 Initial experimental snapshot of next-gen storage.
Change-Id: Ifb8709960dbedd1d9f5efd88cdd359ee9fa9d26d
2014-11-25 17:02:00 +01:00
Julius Volz 85497e3f38 Add function to drop common labels in a vector.
This fixes https://github.com/prometheus/prometheus/issues/384.

Change-Id: I2973c4baeb8a4618ec3875fb11c6fcf5d111784b
2014-11-25 17:02:00 +01:00
Julius Volz 3fdb74e571 Add more topk() / bottomk() tests.
Test what happens if k > number of input elements.

Change-Id: Ie724b850939e297ebf085f0a5a3522e9cfcc6534
2014-11-25 17:02:00 +01:00
Julius Volz c582ae73c2 Implement topk() and bottomk() functions.
To achieve O(log n * k) runtime, this uses a heap to track the current
bottom-k or top-k elements while iterating over the full set of
available elements.

It would be possible to reuse more code between topk and bottomk, but I
decided for some more duplication for the sake of clarity.

This fixes https://github.com/prometheus/prometheus/issues/399

Change-Id: I7487ddaadbe7acb22ca2cf2283ba6e7915f2b336
2014-11-25 17:02:00 +01:00
Julius Volz 00b9489f1c Fix time() behavior.
time() should return the timestamp for which the query is executed, not
the actual current time.

Change-Id: I430a45cabad7785cd58f95b1028a71dff4c87710
2014-11-25 17:02:00 +01:00
Julius Volz c5984f1818 Add abs() and over-time aggregation functions.
This implements aggregation functions over time as request in
https://github.com/prometheus/prometheus/issues/383.

Change-Id: Ifd69b850de8cfdf6e7a6c0e042056fa4c672410e
2014-11-25 17:02:00 +01:00
Brian Brazil e041c0cd46 Add console and alert templates with access to all data.
Move rulemanager to it's own package to break cicrular dependency.
Make NewTestTieredStorage available to tests, remove duplication.

Change-Id: I33b321245a44aa727bfc3614a7c9ae5005b34e03
2014-05-30 16:24:56 +01:00
Bjoern Rabenstein ca6a4fccef Weed out our homegrown test.Tester.
The Go stdlib has testing.TB now, which fulfills the exact same
purpose.

Change-Id: I0db9c73400e208ca376b932a02b7e3402234b87c
2014-05-21 19:27:24 +02:00
Julius Volz 01f652cb4c Separate storage implementation from interfaces.
This was initially motivated by wanting to distribute the rule checker
tool under `tools/rule_checker`. However, this was not possible without
also distributing the LevelDB dynamic libraries because the tool
transitively depended on Levigo:

rule checker -> query layer -> tiered storage layer -> leveldb

This change separates external storage interfaces from the
implementation (tiered storage, leveldb storage, memory storage) by
putting them into separate packages:

- storage/metric: public, implementation-agnostic interfaces
- storage/metric/tiered: tiered storage implementation, including memory
                         and LevelDB storage.

I initially also considered splitting up the implementation into
separate packages for tiered storage, memory storage, and LevelDB
storage, but these are currently so intertwined that it would be another
major project in itself.

The query layers and most other parts of Prometheus now have notion of
the storage implementation anymore and just use whatever implementation
they get passed in via interfaces.

The rule_checker is now a static binary :)

Change-Id: I793bbf631a8648ca31790e7e772ecf9c2b92f7a0
2014-04-16 13:30:19 +02:00
Julius Volz d411a7d810 Allow reversing vector and scalar arguments in binops.
This allows putting a scalar as the first argument of a binary operator
in which the second argument is a vector:

  <scalar> <binop> <vector>

For example,

  1 / http_requests_total

...will output a vector in which every sample value is 1 divided by the
respective input vector element.

This even works for filter binary operators now:

  1 == http_requests_total

Returns a vector with all values set to 1 for every element in
http_requests_total whose initial value was 1.

Note: For filter binary operators, the resulting values are always taken
from the left-hand-side of the operation, no matter whether the scalar
or the vector argument is the left-hand-side. That is,

  1 != http_requests_total

...will set all result vector sample values to 1, although these are
exactly the sample elements that were != 1 in the input vector.

If you want to just filter elements without changing their sample
values, you still need to do:

  http_requests_total != 1

The new filter form is a bit exotic, and so probably won't be used
often. But it was easier to implement it than disallow it completely or
change its behavior.

Change-Id: Idd083f2bd3a1219ba1560cf4ace42f5b82e797a5
2014-04-08 17:16:18 +02:00
Julius Volz c7c0b33d0b Add regex-matching support for labels.
There are four label-matching ops for selecting timeseries now:

- Equal: =
- NotEqual: !=
- RegexMatch: =~
- RegexNoMatch: !~

Instead of looking up labels by a simple clientmodel.LabelSet (basically
an equals op for every key/value pair in the set), timeseries
fingerprint selection is now done via a list of metric.LabelMatchers.

Change-Id: I510a83f761198e80946146770ebb64e4abc3bb96
2014-04-01 14:24:53 +02:00
Julius Volz 7e9ecaac3a Add count_scalar() function.
Change-Id: I63f09dd0479d0a6b016f5f857dd39dcbda56c7f9
2014-01-30 13:07:26 +01:00
Julius Volz 0378c2ca1f Nonexistent labels in BY-clauses shouldn't propagate to result.
This fixes bug 2. of https://github.com/prometheus/prometheus/issues/374

Change-Id: Ia4a13153616bafce5bf10597966b071434422d09
2014-01-24 16:05:30 +01:00
Julius Volz 6dc36d0c3e Don't keep extra labels in aggregations by default.
MIN/MAX/SUM/AVG/COUNT aggregations will now by default drop all labels that are
not specifically part of a BY-clause, even if a label value is the same within
all timeseries of an aggregation group. The old behavior of keeping extra
labels may still be switched on by adding KEEPING_EXTRA to the end of an
aggregation statement:

  sum(http_requests) by (job, method) keeping_extra

I'm open to better syntax/naming suggestions.

Change-Id: I21d3fe7af9e98552ce3dffa3ce7c0a4ba4c0b4a4
2013-12-16 12:53:10 +01:00
Julius Volz 740d448983 Use custom timestamp type for sample timestamps and related code.
So far we've been using Go's native time.Time for anything related to sample
timestamps. Since the range of time.Time is much bigger than what we need, this
has created two problems:

- there could be time.Time values which were out of the range/precision of the
  time type that we persist to disk, therefore causing incorrectly ordered keys.
  One bug caused by this was:

  https://github.com/prometheus/prometheus/issues/367

  It would be good to use a timestamp type that's more closely aligned with
  what the underlying storage supports.

- sizeof(time.Time) is 192, while Prometheus should be ok with a single 64-bit
  Unix timestamp (possibly even a 32-bit one). Since we store samples in large
  numbers, this seriously affects memory usage. Furthermore, copying/working
  with the data will be faster if it's smaller.

*MEMORY USAGE RESULTS*
Initial memory usage comparisons for a running Prometheus with 1 timeseries and
100,000 samples show roughly a 13% decrease in total (VIRT) memory usage. In my
tests, this advantage for some reason decreased a bit the more samples the
timeseries had (to 5-7% for millions of samples). This I can't fully explain,
but perhaps garbage collection issues were involved.

*WHEN TO USE THE NEW TIMESTAMP TYPE*
The new clientmodel.Timestamp type should be used whenever time
calculations are either directly or indirectly related to sample
timestamps.

For example:
- the timestamp of a sample itself
- all kinds of watermarks
- anything that may become or is compared to a sample timestamp (like the timestamp
  passed into Target.Scrape()).

When to still use time.Time:
- for measuring durations/times not related to sample timestamps, like duration
  telemetry exporting, timers that indicate how frequently to execute some
  action, etc.

*NOTE ON OPERATOR OPTIMIZATION TESTS*
We don't use operator optimization code anymore, but it still lives in
the code as dead code. It still has tests, but I couldn't get all of them to
pass with the new timestamp format. I commented out the failing cases for now,
but we should probably remove the dead code soon. I just didn't want to do that
in the same change as this.

Change-Id: I821787414b0debe85c9fffaeb57abd453727af0f
2013-12-03 09:11:28 +01:00
Julius Volz 35ee2cd3cb Add alertmanager notification support to Prometheus.
Alert definitions now also have mandatory SUMMARY and DESCRIPTION fields
that get sent along a firing alert to the alert manager.
2013-07-30 17:23:41 +02:00
Julius Volz 64b0ade171 Swap rules lexer for much faster one.
This swaps github.com/kivikakk/golex for github.com/cznic/golex.

The old lexer would have taken 3.5 years to load a set of 5000 test rules
(quadratic time complexity for input length), whereas this one takes only 32ms.
Furthermore, since the new lexer is embedded differently, this gets rid of the
global parser variables and makes the rule loader fully reentrant without a
lock.
2013-07-11 19:35:29 +02:00
Matt T. Proud 30b1cf80b5 WIP - Snapshot of Moving to Client Model. 2013-06-25 15:52:42 +02:00
Julius Volz fc97e688c6 Improve printing of rules and expressions. 2013-06-11 11:39:31 +02:00
Matt T. Proud 2c3df44af6 Ensure database access waits until it is started.
This commit introduces a channel message to ensure serving
state has been reached with the storage stack before anything attempts
to use it.
2013-06-06 10:42:21 +02:00
Julius Volz 51689d965d Add debug timers to instant and range queries.
This adds timers around several query-relevant code blocks. For now, the
query timer stats are only logged for queries initiated through the UI.
In other cases (rule evaluations), the stats are simply thrown away.

My hope is that this helps us understand where queries spend time,
especially in cases where they sometimes hang for unusual amounts of
time.
2013-06-05 18:32:54 +02:00
Julius Volz 138334fb31 Fix handling of negative deltas for non-counter values. 2013-05-28 17:36:53 +02:00
Julius Volz 750f862d9a Use GetBoundaryValues() for non-counter deltas. 2013-05-22 19:13:47 +02:00
Bernerd Schaefer 63d9988b9c Drop unused writeMemoryInterval 2013-05-14 17:03:03 +02:00
Bernerd Schaefer aa96c7d141 Fix rules_test.go
This is smelly, but for now we copy a helper method from the metric
tests into rules.
2013-05-14 16:55:18 +02:00
Julius Volz 0877680761 Implement a COUNT ... BY aggregation operator.
This also removes the now obsolete scalar count() function and corrects the
expressions test naming (broken in
2202cd71c9 (L6R59))
so that the expression tests will actually run.
2013-05-08 16:35:16 +02:00
Julius Volz 56324d8ce2 Make AST query storage non-global. 2013-05-07 13:15:10 +02:00
Matt T. Proud ce45787dbf Storage interface to TieredStorage.
This commit drops the Storage interface and just replaces it with a
publicized TieredStorage type.  Storage had been anticipated to be
used as a wrapper for testability but just was not used due to
practicality.  Merely overengineered.  My bad.  Anyway, we will
eventually instantiate the TieredStorage dependencies in main.go and
pass them in for more intelligent lifecycle management.

These changes will pave the way for managing the curators without
Law of Demeter violations.
2013-05-03 15:54:14 +02:00
Julius Volz 2202cd71c9 Track alerts over time and write out alert timeseries. 2013-04-26 14:35:21 +02:00
Julius Volz c0601abf46 Implement initial no-op alert parsing and rule parsing tests. 2013-04-23 13:48:24 +02:00
Matt T. Proud f9e99bd08a Refresh SampleValue to 64-bit floating point.
We always knew that this needed to be fixed.
2013-04-21 20:31:50 +02:00
Julius Volz 5f5ea03105 Run "make format". 2013-04-16 17:23:59 +02:00
Julius Volz 1cff4f3d91 Fix rate() per-second adjustment.
This got broken during the depointerization of the Vector type.
2013-04-15 14:41:34 +02:00