Lost a lot of metric data points in the past few hours

Hi, it seems that there is something wrong with the wandb server as I lost a lot of data points in the past few hours. I discussed this with some other users in Discord and it’s not only me!

Hi @geoffrey-chen thanks for reporting it. Could you please provide more information what type of data you’ve lost, is it metrics or artifact files? Which is your current W&B Server/Local version? and are you deploying this locally in your machine using Docker or is it hosted elsewhere?

Hi @geoffrey-chen I wanted to follow up here regarding this issue you reported with lost data points. May I please ask if you were referring to our public cloud wandb.ai or to a self-hosted W&B Server instance?

There was an incident reported 5 days ago: https://status.wandb.com/ where the metrics were delayed in showing up in the UI. This has been fixed by our Infra team. Could you please let us know if that issue still occurs for you?

Hi, my issue has been resolved after a few hours. I guess that incident is the reason.

Thanks

Perfect, glad to hear this @geoffrey-chen and thanks for the confirmation! I will now close this ticket, but please feel free to reach out to us here if you had any other questions/issues.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.