Cortex read latency spikes

Incident Report for Grafana Cloud

Resolved

An issue in the compactor service caused a large backlog of un-compacted blocks to build up in object storage. This caused elevated query latency for customers in cortex and in some cases caused queries to timeout. Writes during this time were successful and no data loss was observed as expected.

Posted Oct 07, 2022 - 18:41 UTC