Request to increase rate limit and missing runs from dashboard

Hi W&B team,

I’m training several models concurrently and am encountering the following error:

wandb: 429 encountered (Filestream rate limit exceeded, retrying in 2.1 seconds.), retrying request

May I kindly request to increase my user rate limit?

Also, I noticed that on our compute cluster, I am running 8 jobs (e.g., checked via squeue), but only six runs appear active in the W&B dashboard. For example, I identified a job, with run name 00:56:03 on 06/30/2024 (29437), that produces logs that confirm that the job is running. The SLURM job ID is 38009820, and I can use squeue to confirm that the job is actively running with runtime 16:31:20. However, I don’t see this run on my W&B dashboard. May you please advise how to resolve this issue?

Thanks!

Hi Ayush,

I will look into this further and see about increasing your rate limits. I’ll get back to you as soon as possible!

Best regards,
Jason

Hi Ayush,

I have increased the RL for you. Please let me know if there is anything else at all I can do for you :slightly_smiling_face:

Best regards,
Jason