* Wiped out a ton of warnings about comments.
* Created consts for `MetricType` values.
* `externalInfoMap` can now track multiple series with the same name/namespace and different labels.
* `namespace` parameter of external metrics queries is now respected (albeit very rudimentarily)
* Metric converter values for external metrics are now converted the same way as for custom metrics (probably still some opportunity for consolidation).
* Lots of TODOs actually done.
* Deleted a lot of commented out code.
Fixed:
* `basicMetricLister` wasn't applying the appropriate start time because I had forgotten to set the `lookback`.
There are still a number of issues:
* The `externalPrometheusProvider` is not hooked up to the web application yet, so it doesn't serve requests.
* The namespace and label approach used in `external_info_map.go` is horrifically incorrect. It doesn't appropriately store multiple series with the same name but different labels.
* The configuration is still not updated to appropriately handle external metrics, it's sort of half-piggy-backing on the pre-existing work.
* Some bug fixes found during testing/test repair.
* Trying to tease apart the various responsibilities of `metricNamer` into smaller chunks and adding tests for each individual chunk.
* Updating the `provider_test` and `series_registry_test` to fix failures.
* Breaking out some types to make the functionality more composable and easier to digest. (e.g. `basicMetricLister` interacts with Prometheus and `periodicMetricLister` periodically invokes `basicMetricLister`.
* Pulling out some of the type embedding between `basicSeriesRegistry` and `MetricLister` to make it easier to digest.
* Deleting the `/metric-converter` code because I'm pretty certain it's not going to be necessary as things transition to using the namer-based configuration.
* Some light-ish refactoring in `metricNamer` to get some re-use out of query generation in preparation for using it with external metrics.
Notable next steps include:
* Intelligently select the aggregation method (rate/avg_over_time/etc).
* Intelligenty select the aggregation window.
* Figure out how to handle "namespace", if at all.
* Fix the crazy type conversion in `sample_converter.go`.
Working through some of the result conversions.
I want to use this code in a harness project to exercise some of it, but there are some versioning issues around my modified code vs. the current version of the real project. The easiest way to work around them at the moment is to push this.
This fixes asynchronous read/write issues to when performing series
discovery by pushing series results onto a channel, instead of trying to
write them directly to a map.
This commit introduces advanced configuration. The rate-interval and
label-prefix flags are removed, and replaced by a configuration file
that allows you to specify series queries and the rules for transforming
those into metrics queries and API resources.
This allows setting a prefix on the labels used to determine which
resources a series belongs to. The prefix may be set using the
`--label-prefix` flag.
This updates the dependencies to be compatible with Kubernetes 1.8,
including the new custom.metrics.k8s.io/v1beta1 API group and the
corresponding custom-metrics-apiserver changes.
This commit switches to using the boilerplate versions of
a couple different utilities, including the metric info normalization
and the common error types.
All errors returned by the provider should be proper Kube API status
errors. This commit cleans up a couple spots that either returned an
invalid error, or were missing an error (such as the case in when no
results matched the requested object).
Previously, if we encountered an error while trying to update our series
list, we'd return an error, aborting the processing of the entire batch.
This could lead to the list of available metrics being severely out of
date. Instead, we simply log an error when we fail to process a metric
name, and skip it.
This causes the Prometheus provider to take a stop channel as an
argument, which allows us to stop the lister (which keeps the series list
up to date) in the unit tests.
Several of the unit tests were out of date with the interface or type
definitions. Now that we're running CI, they need to be up to date
(they should have been anyway, but that's a different story).
k8s.io/custom-metrics-boilerplate was "renamed" to
"github.com/directxman12/custom-metrics-boilerplate". This updates all
references to point to the correct location.
The initial functionality works. There's still a number of TODOs to
clean up, and some edge cases to work around, and some errors that could
be handled better.
