Failed to verify certificate

W&B Help
1

Hi, I am running my code on a remote server that may someone else is also running its job and use the wandb.

This is how I use wandb with pytorch lightning

API_TOKEN_WANDB = "<my api token>"
os.environ["WANDB_API_KEY"] = API_TOKEN_WANDB
wandb_logger = WandbLogger(save_dir=args.log_dir, project="Project", name="Test")

trainer = pl.Trainer(
        logger=wandb_logger,
        accelerator=compute["accelerator"],
        devices=compute["devices"],
        min_epochs=1,
        max_epochs=args.epochs,
        precision=compute["precision"],
        callbacks=callbacks,
    )

And here the error I face every time:

wandb.errors.CommError: Failed to get resume status for run owqrvoj9: api: failed sending: POST https://api.wandb.ai/graphql giving up after 1 attempt(s): Post "https://api.wandb.ai/graphql": tls: failed to verify certificate: x509: certificate signed by unknown authority
INFO:    Terminating fuse-overlayfs after timeout
INFO:    Timeouts can be caused by a running background process

Just for your notice that others can use wandb but I can’t in the remote server.

2

Hi @elsobhano Good day and thank you for reaching out to us. The error message appears to be some permission issues on your remote server. Are you able to run your code on this remote server before? Also, are you also a member of the Team where other users are part of?

3

Hi @elsobhano , since we have not heard back from you we are going to close this request. If you would like to re-open the conversation, please let us know!

4

Yes, I am a member of a team. I thinks the server has its own policies, and work with https:// well. But I don’t know what is wrong with it

5

At least you could explain what is going wrong here and how to fix it.