Some elevated error rates on the prometheus-blocks-prod-us-central1, influx-blocks-prod-us-central1 and graphite-blocks-prod-us-central1 clusters.

Resolved

We continue to observe a period of recovery. At this time, we are considering this issue resolved.

Update

We have observed marked improvement with the issue. Customers should no longer experience this issue. We will continue to monitor and provide updates.

Investigating

We are continuing to investigate this issue.

Investigating

It seems that the cloud provider object storage has latency spikes, which directly affect our write latency.