Michael Hoffmann
Michael Hoffmann
Added failing test here: https://github.com/thanos-io/thanos/pull/7468 This Issue is to track the work of fixing that test.
This PR makes sidecar read "minT" from prometheus metrics instead of the shipper. This has the benefit that it also works for sidecar deployments that do not upload blocks, so...
* [ ] I added CHANGELOG entry for this change. * [ ] Change is not relevant to the end user. ## Changes ## Verification
### Is your proposal related to a problem? Downsampling right now does not support native histograms. We should think about how a downsampled native histogram chunk should look and adapt...
If we have a Querier that queries a Sidecar and a Storage Gateway that the Sidecar uploads to and if we query a range of data that could be completely...
Fixes #56 We give index, splat and get_attr higher precedence then unary operations so that `!foo.bar` reads as `!(foo.bar)` instead of `(!foo).bar`
This commit removes the annotation that is added when a metric has a name that is not indicative of a counter. This annotation is mostly spam that the query author...
This PR fixes a bug in ts_of_last_over_time where the float samples where used when computing the last timestamp of the histogram samples. Signed-off-by: Michael Hoffmann
`topk(NaN, non_existent)` used to return an error, it stopped after https://github.com/prometheus/prometheus/pull/16404/files. This is a minor issue, but broke thanos regression tests
We add a "service_config" field to endpoint config file that we can use to override the default service_config for endpoint groups. This enables us to configure retry policy or loadbalncing...