New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Cannot get api key after run local WANDB #1054
Comments
Got the same problem today after upgrading wandb-local with |
This looks like the migration failed on restart. Can you run the following and attach the input.
You can also attempt manual restarting, and share the logs from first boot:
|
@stalkermustang your migration failed at version 8 which is suspect. CAn you also run:
From within the mysql shell? There should be an "agent_id" column. Assuming there is, you can run the following from the mysql shell:
Then restart the instance to have new migrations applied. |
@vanpelt |
also i cannot delete users from |
so i'm remove image, stop and remove container, then rm volume wandb from docker. Also i clear wandb file on my local server (with log and settings).
what magic is it? @vanpelt |
Now log be like:
|
Another detailed try:
^- at this point i don't see /authorize link like first time
and output is:
(version 24 what???)
i got:
agent_id here, huh
|
have no idea what is going... |
@stalkermustang when the container boots it will run migrations. If you kill or stop the container during the running of these migrations you'll end up with the database in an invalid state. We need to get to the bottom of why your migration is failing. Can you run the following?
Once this is up and running you can see the migration progress in the vm by querying the database:
You'd will need to run the SELECT statement multiple times and ultimately it should return finish at version 67 with the dirty column equal to 0. |
@stalkermustang the data exists in a docker managed volume named "wandb" which I removed above by running |
I am getting simliar problem, may I know how can I fix it? |
I waited an hour until my migration ended (I expected the process to be much faster).
at this point i go to /authorize page and ... done! |
This is likely due to the docker file system driver. @noklam @stalkermustang what OS are you running docker from? Switching to the "overlay" filesystem would likely make this run much faster. |
I am running from Window. |
Can you run |
|
Weights and Biases version: version 0.8.36
Python version: 3.7.5
Operating System: Linux
Description
I'm running wandb localy:
wandb local --upgrade --port 8892
Then i get in console:
wandb: You can find your API key in your browser here: http://localhost:8892/authorize wandb: Paste an API key from your profile and hit enter:
And when i go to page, i see
this endless (and reloading) message. Sometimes browser redirect me to /signup page, but there's error:
restarts and rebuilds didn't help.
In debug.log i only see:
2020-05-21 17:03:16,408 DEBUG MainThread:22316 [auth.py:find_config_file():47] Trying paths: ['/home/seeall/.docker/config.json', '/home/seeall/.dockercfg'] 2020-05-21 17:03:16,408 DEBUG MainThread:22316 [auth.py:find_config_file():54] No config file found 2020-05-21 17:03:16,598 DEBUG MainThread:22316 [auth.py:resolve_authconfig():295] Looking for auth entry for 'docker.io' 2020-05-21 17:03:16,599 DEBUG MainThread:22316 [auth.py:resolve_authconfig():306] No entry found 2020-05-21 17:03:19,564 DEBUG MainThread:22316 [auth.py:resolve_authconfig():295] Looking for auth entry for 'docker.io' 2020-05-21 17:03:19,564 DEBUG MainThread:22316 [auth.py:resolve_authconfig():306] No entry found
BUT dockerlogs spam:
The text was updated successfully, but these errors were encountered: