promqltest: Complete the tests for info annotations
So far, we did not test for the _absence_ of an info annotation
(because many tests triggered info annotations, which we haven't taken
into account so far).
The test for info annotations was also missed for range queries.
This completes the tests for info annotations (and refactors the many
`if` statements into a somewhat more compact `switch` statement).
It fixes most tests to not emit an info annotation anymore. Or it
changes the `eval` to `eval_info` where we actually want to test for
the info annotation.
It also fixes a few spelling errors in comments.
---------
Signed-off-by: beorn7 <beorn@grafana.com>
Signed-off-by: Björn Rabenstein <github@rabenste.in>
Co-authored-by: Arve Knudsen <arve.knudsen@gmail.com>
This converts `TestNativeHistogram_SubOperator` to the promql testing framework. It also removes `TestNativeHistogram_Sum_Count_Add_AvgOperator`, which got converted earlier.
Signed-off-by: Neeraj Gartia <neerajgartia211002@gmail.com>
* tests(promql/testdata): add regression test for and-on
I'd like to use queries of the form "x and on() (vector(y)==1)" to be
able to include and exclude series for dashboards. This helps migration
to native histograms in dashboards by using a dashboard variable to
set "y" to either -1 or 1 to exclude or include the result.
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
---------
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
* Fix issue where comparison operations with `bool` modifier and native histograms return histograms rather than 0 or 1
* Don't emit anything for comparisons between floats and histograms when `bool` modifier is used
* Don't emit anything for comparisons between floats and histograms when `bool` modifier is used between a vector and a scalar
---------
Signed-off-by: Charles Korn <charles.korn@grafana.com>
PromQL: Correct the behaviour of some operator and aggregators with Native Histograms
---------
Signed-off-by: Neeraj Gartia <neerajgartia211002@gmail.com>
In general aim for the happy case when the exposer lists the buckets
in ascending order.
Use Compact(2) to compact the result of nhcb convert.
This is more in line with how client_golang optimizes spans vs
buckets.
aef8aedb4b/prometheus/histogram.go (L1485)
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.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>
promql: Fix stddev/stdvar when aggregating histograms, NaNs, and Infs
Native histograms are ignored when calculating stddev or stdvar.
However, for the first series of each group, a `groupedAggregation` is
always created. If the first series that was encountered is a histogram
then it acts as the equivalent of a 0 point.
This change creates the first `groupedAggregation` with the `seen` field set to `false` if the point is a
histogram, thus ignoring it like the rest of the aggregation function does. A new `groupedAggregation`
will then be created once an actual float value is encountered.
This commit also sets the `floatValue` field of the `groupedAggregation` to `NaN`, if the first
float value of a group is `NaN` or `±Inf`, so that the outcome is consistently `NaN` once those
values are in the mix.
(The added tests fail without this change).
Signed-off-by: Joshua Hesketh <josh@nitrotech.org>
Signed-off-by: beorn7 <beorn@grafana.com>
---------
Signed-off-by: Joshua Hesketh <josh@nitrotech.org>
Signed-off-by: beorn7 <beorn@grafana.com>
Co-authored-by: beorn7 <beorn@grafana.com>
The `info` function is an experiment to improve UX
around including labels from info metrics.
`info` has to be enabled via the feature flag `--enable-feature=promql-experimental-functions`.
This MVP of info simplifies the implementation by assuming:
* Only support for the target_info metric
* That target_info's identifying labels are job and instance
Also:
* Encode info samples' original timestamp as sample value
* Deduce info series select hints from top-most VectorSelector
---------
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
Co-authored-by: Ying WANG <ying.wang@grafana.com>
Co-authored-by: Augustin Husson <augustin.husson@amadeus.com>
Co-authored-by: Bartlomiej Plotka <bwplotka@gmail.com>
Co-authored-by: Björn Rabenstein <github@rabenste.in>
Co-authored-by: Bryan Boreham <bjboreham@gmail.com>
promql: corrects binary operators functioning for mixed sample with histogram and float
For invalid pairings of sample types, an annotation is added now.
Signed-off-by: Neeraj Gartia <neerajgartia211002@gmail.com>
---------
Signed-off-by: Neeraj Gartia <neerajgartia211002@gmail.com>
* model: move classic to NHCB conversion into its own file
In preparation for #14978.
Author: Jeanette Tan <jeanette.tan@grafana.com> 2024-07-03 11:56:48
Signed-off-by: Jeanette Tan <jeanette.tan@grafana.com>
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
Co-authored-by: Jeanette Tan <jeanette.tan@grafana.com>
Co-authored-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
* Better naming from review comment
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
* Add doc strings.
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
---------
Signed-off-by: Jeanette Tan <jeanette.tan@grafana.com>
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
Co-authored-by: Jeanette Tan <jeanette.tan@grafana.com>
For: #14355
This commit updates Prometheus to adopt stdlib's log/slog package in
favor of go-kit/log. As part of converting to use slog, several other
related changes are required to get prometheus working, including:
- removed unused logging util func `RateLimit()`
- forward ported the util/logging/Deduper logging by implementing a small custom slog.Handler that does the deduping before chaining log calls to the underlying real slog.Logger
- move some of the json file logging functionality to use prom/common package functionality
- refactored some of the new json file logging for scraping
- changes to promql.QueryLogger interface to swap out logging methods for relevant slog sugar wrappers
- updated lots of tests that used/replicated custom logging functionality, attempting to keep the logical goal of the tests consistent after the transition
- added a healthy amount of `if logger == nil { $makeLogger }` type conditional checks amongst various functions where none were provided -- old code that used the go-kit/log.Logger interface had several places where there were nil references when trying to use functions like `With()` to add keyvals on the new *slog.Logger type
Signed-off-by: TJ Hoplock <t.hoplock@gmail.com>
These functions operate on whole series, not on samples, so they do not
fit into the table of functions that return a Vector. Remove the stub
entries that were left to help downstream users of the code identify
what changed.
We cannot remove the entries from the `FunctionCalls` map without
breaking `TestFunctionList`, so put some nils in to keep it happy.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
Go's sorting functions can re-order equal elements, so the strategy of
sorting by the fallback ordering first does not always work.
Pulling the fallback into the main comparison function is more reliable
and more efficient.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
* PromQL.Engine: Refactor Matrix expansion into a method
Add utility method promql.evaluator.expandSeriesToMatrix, for expanding a slice
of storage.Series into a promql.Matrix.
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
* Rename to generateMatrix
Rename evaluator.expandSeriesToMatrix into generateMatrix, while also dropping
the start, end, interval arguments since they are evaluator fields.
Write more extensive method documentation.
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
* Rename to evalVectorSelector
Rename to evalVectorSelector after discussing with @michahoffmann.
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
---------
Signed-off-by: Arve Knudsen <arve.knudsen@gmail.com>
promql: correctly handle unary negation of native histograms and add tests for multiplication and division of native histograms by negative scalars
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
The linear interpolation (assuming that observations are uniformly
distributed within a bucket) is a solid and simple assumption in lack
of any other information. However, the exponential bucketing used by
standard schemas of native histograms has been chosen to cover the
whole range of observations in a way that bucket populations are
spread out over buckets in a reasonably way for typical distributions
encountered in real-world scenarios.
This is the origin of the idea implemented here: If we divide a given
bucket into two (or more) smaller exponential buckets, we "most
naturally" expect that the samples in the original buckets will split
among those smaller buckets in a more or less uniform fashion. With
this assumption, we end up with an "exponential interpolation", which
therefore appears to be a better match for histograms with exponential
bucketing.
This commit leaves the linear interpolation in place for NHCB, but
changes the interpolation for exponential native histograms to
exponential. This affects `histogram_quantile` and
`histogram_fraction` (because the latter is more or less the inverse
of the former).
The zero bucket has to be treated specially because the assumption
above would lead to an "interpolation to zero" (the bucket density
approaches infinity around zero, and with the postulated uniform usage
of buckets, we would end up with an estimate of zero for all quantiles
ending up in the zero bucket). We simply fall back to linear
interpolation within the zero bucket.
At the same time, this commit makes the call to stick with the
assumption that the zero bucket only contains positive observations
for native histograms without negative buckets (and vice versa). (This
is an assumption relevant for interpolation. It is a mostly academic
point, as the zero bucket is supposed to be very small anyway.
However, in cases where it _is_ relevantly broad, the assumption helps
a lot in practice.)
This commit also updates and completes the documentation to match both
details about interpolation.
As a more high level note: The approach here attempts to strike a
balance between a more simplistic approach without any assumption, and
a more involved approach with more sophisticated assumptions. I will
shortly describe both for reference:
The "zero assumption" approach would be to not interpolate at all, but
_always_ return the harmonic mean of the bucket boundaries of the
bucket the quantile ends up in. This has the advantage of minimizing
the maximum possible relative error of the quantile estimation.
(Depending on the exact definition of the relative error of an
estimation, there is also an argument to return the arithmetic mean of
the bucket boundaries.) While limiting the maximum possible relative
error is a good property, this approach would throw away the
information if a quantile is closer to the upper or lower end of the
population within a bucket. This can be valuable trending information
in a dashboard. With any kind of interpolation, the maximum possible
error of a quantile estimation increases to the full width of a bucket
(i.e. it more than doubles for the harmonic mean approach, and
precisely doubles for the arithmetic mean approach). However, in
return the _expectation value_ of the error decreases. The increase of
the theoretical maximum only has practical relevance for pathologic
distributions. For example, if there are thousand observations within
a bucket, they could _all_ be at the upper bound of the bucket. If the
quantile calculation picks the 1st observation in the bucket as the
relevant one, an interpolation will yield a value close to the lower
bucket boundary, while the true quantile value is close to the upper
boundary.
The "fancy interpolation" approach would be one that analyses the
_actual_ distribution of samples in the histogram. A lot of statistics
could be applied based on the information we have available in the
histogram. This would include the population of neighboring (or even
all) buckets in the histogram. In general, the resolution of a native
histogram should be quite high, and therefore, those "fancy"
approaches would increase the computational cost quite a bit with very
little practical benefits (i.e. just tiny corrections of the estimated
quantile value). The results are also much harder to reason with.
Signed-off-by: beorn7 <beorn@grafana.com>
Neeraj Gartia