Resolved -
This incident has been resolved.
Jun 10, 00:08 PDT
Monitoring -
We identified an infrastructure issue as the root cause and have resolved it. We are rapidly working through the backlog of ingested metrics and should fully catch up soon.
Jun 9, 22:44 PDT
Identified -
We are currently responding to delay of up to 5 hours for metrics written into wandb. We have identified the issue and are working to process metrics as quickly as possible. There is no data loss and all data will be complete once the backlog has drained. We're very sorry for the disruption.
Jun 9, 16:47 PDT