Resolved -
We've identified the root cause causing the OOM issues on our ClickHouse cluster and have mitigated the issue. We've replaced cluster nodes that were lost during the incident. All systems are operating as expected!
Sep 28, 03:47 UTC
Investigating -
We've spotted that something has gone wrong with memory consumption in ClickHouse. We're currently investigating the issue, and will provide an update soon.
In the meantime queries may return more slowly or not at all.
Sep 27, 10:45 UTC