Commit graph

220 commits

Author SHA1 Message Date
Fabian Reinartz 5e13880201 General cleanup of rules. 2015-06-06 21:40:52 +02:00
Fabian Reinartz 75c920c95e Remove DotGraph method from Rule interface 2015-06-06 21:35:59 +02:00
Fabian Reinartz 83d07516e8 Remove EvalRaw methods from Rule interface 2015-06-06 21:34:09 +02:00
Fabian Reinartz 280d11dca8 main: exit on invalid rule files on startup. 2015-06-02 18:44:41 +02:00
Fabian Reinartz 0de6edbdfc Move pkg/ to util/ 2015-06-01 21:12:32 +02:00
Fabian Reinartz 02717e6fde Remove generic set type 2015-06-01 21:12:32 +02:00
Fabian Reinartz dbc0d30e3e Move string functionality to pkg/strutil 2015-06-01 21:12:32 +02:00
Fabian Reinartz f45a5cab60 Move templates package to pkg/template 2015-06-01 21:12:31 +02:00
Fabian Reinartz c44ac7bc26 Load rule files from entire directories 2015-06-01 21:12:31 +02:00
Julius Volz d7c015c149 Convert pathPrefix to not have trailing slash. 2015-06-01 12:43:17 +02:00
Julius Volz ff53d10849 Fix double slash in GeneratorURL sent to alertmanager.
Fixes https://github.com/prometheus/prometheus/issues/722
2015-05-23 19:16:57 +02:00
Julius Volz 267fd34156 Switch Prometheus to use github.com/prometheus/log.
This change is conceptually very simple, although the diff is large. It
switches logging from "github.com/golang/glog" to
"github.com/prometheus/log", while not actually changing any log
messages. V(1)-style logging has been changed to be log.Debug*().
2015-05-20 18:19:32 +02:00
Fabian Reinartz e2ed921505 Merge branch 'master' into fabxc/servdisc 2015-05-20 14:13:08 +02:00
Mitsuhiro Tanda 3e914a8cb1 fix graph links with path prefix 2015-05-19 02:45:05 +09:00
Fabian Reinartz bb540fd9fd Implement config reloading on SIGHUP.
With this commit, sending SIGHUP to the Prometheus process will reload
and apply the configuration file. The different components attempt
to handle failing changes gracefully.
2015-05-13 16:49:46 +02:00
Fabian Reinartz fe935179cd Stop routing rule statements through the engine. 2015-04-29 18:01:43 +02:00
Fabian Reinartz 8d7c479fed Merge pull request #658 from prometheus/fabxc/pql/rules-manager
Rename RuleManager to Manager, remove interface.
2015-04-29 16:54:21 +02:00
Fabian Reinartz 479891c9be Rename RuleManager to Manager, remove interface.
This commits renames the RuleManager to Manager as the package
name is 'rules' now. The unused layer of abstraction of the
RuleManager interface is removed.
2015-04-29 16:42:10 +02:00
Fabian Reinartz 25cdff3527 Remove name arg from Parse* functions, enhance parsing errors. 2015-04-29 16:38:41 +02:00
Fabian Reinartz 3ca11bcaf5 Switch Prometheus to promql package.
This commit removes all functionality from rules/ that is now handled in
promql/.
All parts of Prometheus are changed to use the promql/ package.
2015-04-28 16:19:23 +02:00
Ceesjan Luiten 0e18784c64 Make all paths absolute to support proxies 2015-04-02 20:36:47 +02:00
Brian Brazil 941f585164 Avoid +InfYs and similar, just display +Inf. 2015-03-28 18:51:41 +00:00
beorn7 a075900f9a Merge branch 'beorn7/persistence' into beorn7/ingestion-tweaks 2015-03-18 19:09:31 +01:00
Fabian Reinartz 624f27f4b6 Add ln, log2, log10 and exp functions to the query language. 2015-03-16 18:26:19 +01:00
Julius Volz b2651027fc Fix special value handling in division and modulo.
This fixes https://github.com/prometheus/prometheus/issues/597
2015-03-16 14:23:40 +01:00
beorn7 be11cb2b07 Remove the sample ingestion channel.
The one central sample ingestion channel has caused a variety of
trouble. This commit removes it. Targets and rule evaluation call an
Append method directly now. To incorporate multiple storage backends
(like OpenTSDB), storage.Tee forks the Append into two different
appenders.

Note that the tsdb queue manager had its own queue anyway. It was a
queue after a queue... Much queue, so overhead...

Targets have their own little buffer (implemented as a channel) to
avoid stalling during an http scrape. But a new scrape will only be
started once the old one is fully ingested.

The contraption of three pipelined ingesters was removed. A Target is
an ingester itself now. Despite more logic in Target, things should be
less confusing now.

Also, remove lint and vet warnings in ast.go.
2015-03-15 14:08:22 +01:00
beorn7 13fcf1ddbc Implement double-delta encoded chunks. 2015-03-05 20:33:26 +01:00
beorn7 9e85ab0eef Apply the new signature/fingerprinting functions from client_golang.
This requires the new version of client_golang (vendoring will follow
in the next commit), which changes the fingerprinting for
clientmodel.Metric.
2015-03-03 18:34:01 +01:00
Fabian Reinartz 182de6b99f Fix unary +/- expressions.
Unary expressions cause parsing errors if they are done in the lexer
by tokenizing them into the number.
This fix moves unary expressions to the parser.
2015-03-03 13:30:08 +01:00
Fabian Reinartz 6f754073d5 Add OR operation and vector matching options.
This commits implements the OR operation between two vectors.
Vector matching using the ON clause is added to limit the set of
labels that define a match between two elements. Group modifiers
(GROUP_LEFT/GROUP_RIGHT) to request many-to-one matching are added.
2015-03-03 11:35:10 +01:00
Julius Volz 0ac931aed1 Also support parsing float formats like "2.". 2015-03-02 12:58:05 +01:00
Julius Volz c2ab54e9a6 Support scientific notation and special float values.
This adds support for scientific notation in the expression language, as
well as for all possible literal forms of +Inf/-Inf/NaN.

TODO: Keep enough state in the parser/lexer to distinguish contexts in
which "Inf", "NaN", etc. should be parsed as a number vs. parsed as a
label name. Currently, foo{nan="bar"} would be a syntax error. However,
that is an existing bug for all our reserved words. E.g. foo{sum="bar"}
is a syntax error as well. This should be fixed separately.
2015-03-01 19:31:16 +01:00
beorn7 1a61bcae07 Fix plural of 'histogram'.
Actually, 'histogram' is Ancient Greek and 3rd declension... ;-)
2015-02-23 15:29:26 +01:00
beorn7 17443d288b Avoid copying of the COWMetric if we already have the metric available. 2015-02-22 01:04:52 +01:00
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 7fefccd929 Write() directly into hash and use model.SeparatorByte. 2015-02-21 17:19:13 +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 67e20acc6c Lower-case some package-internal names. 2015-02-18 02:45:54 +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
Julius Volz 82613527f3 Remove unnecessary float64() conversion in round(). 2015-02-05 15:14:05 +01:00
Marko Mikulicic 8fdacbdf17 Add floor, ceil and round functions. Closes #402 2015-02-04 17:20:56 +01:00
Fabian Reinartz fa1e90003b Query timeout added.
This is related to #454. Queries now timeout after a duration set by
the -query.timeout flag. The TotalEvalTimer is now started/stopped
inside any of the ast.Eval* functions.
2015-02-03 08:04:27 +01:00
Bjoern Rabenstein 26e22e6ad6 Fix rule manager shutdown. 2015-01-29 15:05:10 +01:00
Julius Volz d4374a9265 More efficient JSON query result format.
This depends on https://github.com/prometheus/client_golang/pull/51.

For vectors, the result format looks like this:

```json
{
   "version": 1,
   "type" : "vector",
   "value" : [
      {
         "timestamp" : 1421765411.045,
         "value" : "65.475000",
         "metric" : {
            "quantile" : "0.5",
            "instance" : "http://localhost:9090/metrics",
            "job" : "prometheus",
            "__name__" : "http_request_duration_microseconds",
            "handler" : "/static/",
            "method" : "get",
            "code" : "304"
         }
      },
      {
         "timestamp" : 1421765411.045,
         "value" : "5826.339000",
         "metric" : {
            "quantile" : "0.9",
            "instance" : "http://localhost:9090/metrics",
            "job" : "prometheus",
            "__name__" : "http_request_duration_microseconds",
            "handler" : "prometheus",
            "method" : "get",
            "code" : "200"
         }
      },
      /* ... */
   ]
}
```

For matrices, it looks like this:

```json
{
   "version": 1,
   "type" : "matrix",
   "value" : [
      {
         "metric" : {
            "quantile" : "0.99",
            "instance" : "http://localhost:9090/metrics",
            "job" : "prometheus",
            "__name__" : "http_request_duration_microseconds",
            "handler" : "/static/",
            "method" : "get",
            "code" : "200"
         },
         "values" : [
            [
               1421765547.659,
               "29162.953000"
            ],
            [
               1421765548.659,
               "29162.953000"
            ],
            [
               1421765549.659,
               "29162.953000"
            ],
            /* ... */
         ]
      }
   ]
}
```
2015-01-26 13:06:22 +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
Bjoern Rabenstein b09453af1d Adjust to new client_golang API. 2015-01-21 15:42:25 +01:00