Marco Pracucci

Querier tracks cortex_request_duration_seconds_count tracks status_code=503 on "context canceled"

3

I'm investigating the alert `MimirRequestErrors` which fired for a short period on a Mimir cluster because the **querier** tracked `cortex_request_duration_seconds_count` metric with `status_code="503"` and `route="prometheus_api_v1_query_range"`. Looking at querier logs I...

Store-gateway: Series() response latency impacted by Send() more than I expected

5

While investigating https://github.com/grafana/mimir/issues/3940 I noticed that the store-gateway `Series()` response latency is impacted by the [`Send()`](https://github.com/grafana/mimir/blob/1a8c6c7a47e57712f8249caadd1424ab88efe922/pkg/storegateway/bucket.go#L963) more than I expected. The following screenshot shows the breakdown of `Series()` latency in...

Store-gateway: add streaming support to LabelNames() API

2

Similarly to https://github.com/grafana/mimir/issues/3348 we should add streaming support to `LabelNames()` API. _This is lower priority than [`LabelValues()`](https://github.com/grafana/mimir/issues/3750)._

Store-gateway: add streaming support to LabelValues() API

1

Similarly to https://github.com/grafana/mimir/issues/3348 we should add streaming support to `LabelValues()` API. ## What's the priority between `LabelValues()` and `LabelNames()`? I think making `LabelValues()` streaming based is higher priority than `LabelNames()`....

Standardise ring HTTP UI between Mimir components using it as client

4

Once https://github.com/grafana/mimir/pull/2957 will be merged, Mimir will have 3 rings used by multiple components: - Ingesters ring - Store-gateway ring - Query-scheduler ring Currently, all the rings UI is exposed...

Store-gateway: inefficient chunks fetching and caching

2

I analysed the bytes touched vs fetched in some of our production Mimir clusters over the last 24h. I used the metrics `cortex_bucket_store_series_data_size_touched_bytes_sum` and `cortex_bucket_store_series_data_size_fetched_bytes_sum`, and got the following (numbers...

Improve "no org id" error

5

When Mimir receives a request without the `X-Scope-OrgID`, it returns the "no org id" error, coming from `weaveworks/common`: https://github.com/grafana/mimir/blob/4a5df428f680588101bd1946c8b6e780cf86e75f/vendor/github.com/weaveworks/common/user/id.go#L19 However, a Mimir user has no idea what the "org id"...

#### What this PR does This PR is a follow up of #4330. In this PR I'm removing the non-streaming support from the store-gateway `Series()`. I've also removed the option...

Add memory pooling for chunks fetched from cache when fine-grained caching is enabled

2

#### What this PR does We've just rolled out the fine-grained chunks fetching and caching in zone-a of a dev cluster and all metrics are better than other 2 zones...

A customer is running a benchmark on Mimir with OOO ingestion enabled. An ingester experienced an issue while opening TSDB at startup. I've got the following log: ``` level=error ts=2023-02-03T07:00:07.966750744Z...