api {
# Dror Lederman’s workspace
web_server: https://app.community.clear.ml
api_server: https://api.community.clear.ml
files_server: https://files.community.clear.ml
credentials {
“access_key” = “SMBEQP52835QTKVQ9BUW”
“secret_key” = “kogXXX”
}
}
(I’m using macos and clearml was used to work by me + it is working on “my” gcp)
Problem solved for me (with no actions on my side)
BTW- I understand that usually we should be able to use standard images for the autoscaler?
yes, it’s a custom image..
2 already done. Will try 1
Using app.clear.ml
Sorry- my mistake- when I open a new autoscaler, the field is not empty
can you please write the full path to the default machine image?
see attached after added permission. I’m getting the same error (I cloned the autoscaler and task)
Me too. But it works through my phone
Should I use machine image for autoscaler or image (not sure what is the difference)?
@<1523701118159294464:profile|ExasperatedCrab78> Thanks. I tried to check it. Can you refer me to the right place to check/ add permission? (I followed the instruction on clearml site with no luck)
This is where I’m getting the error (“Error: could not verify credentials:“)
Yes, I did. When I install clearml, it doesn’t generate this conf file
so- first I didn’t have any such file. Shouldn’t this file be generated by the initialization procedure?
AgitatedDove14 I’m not sure why, but now I got home and it works (when I wrote you, I tried it from the office; no firewall in the office)
Anyway, thanks a lot for your help
AgitatedDove14 Detected credentials key=“UXVWG8IKGJBFKT8PHDW7” secret=“VHKs***”
WEB Host configured to: [ https://app.community.clear.ml ]
API Host configured to: [ https://api.community.clear.ml ]
File Store Host configured to: [ https://files.community.clear.ml ]
ClearML Hosts configuration:
Web App: https://app.community.clear.ml
API: https://api.community.clear.ml
File Store: https://files.community.clear.ml
Verifying credentials ...
Error: could not verify credentials: key=UX...