Reputation
Badges 1
53 × Eureka!because kubectl inside pod uses inpod method
Hi @<1523701717097517056:profile|ScantMoth28> , disabling atm we are not supporting Istio but I’m more than willing to look at a proposal like yours. Let’s discuss this on a new issue on github pls so we can keep track of it and find a good way to implement. thanks
SuccessfulKoala55 after looking at the issue I’m a bit confused 😄 ; as far as I can see there is no way to pass any parameter to clearml-agent in daemon mode to push log to stdout. Can you confirm it? (If yes I need to find some workaround)
ok, will try to find a solution then, ty
trying a new one from scratch
Did you use any override values.yaml file for this installation ?
then you should go in UI creating a new pair of access/secretkey and use them as parameters for agent so it can connect to clearml and authenticate
Ok, what I need is the exact version of the chart (not app) for clearml and clearml-agent
btw I think it's related clearml-task
usage; did you try to create a pipeline with decorators creating it with clearml-task
before?
I need to investigate, ScrawnyLion96 can you pls open an issue on https://github.com/allegroai/clearml-helm-charts ?
https://artifacthub.io/packages/helm/allegroai/clearml and click on install button so you can see details about repo
I didn’t get the point, if you need to change it, you can just override image name and tag by Helm
as usual it starts small and after 5 mins discussion is getting challenging 😄 I love this stuff... let me think a bit about it I will get back to you asap on this.
apiserver additionalConfigs
moreover I usually prefer to have S3 backends if in AWS oir MinIO
but you are starting from major for that is really old and where naming was potentially unconsistent some time so, my suggestion is to backup EVERY pv before proceeding
really weird, can you try to totally remove any cookie domain related?
ok got it, are you able to access the system bypassing nginx with http://<Server Address>:8080
?
how you cluster reacts is about scaling infra as much as needed (karpenter or any other cloud autoscaler should work)
about autoscaling it’s a complex topic regarding platform management in this case. ClearML glue simply spawn pods with resources defined in template.
@<1673863788857659392:profile|HomelyRabbit25> I see your comment on that PR, if no further feedback will be there, I will try to eventually improve that PR by myself when possible.
There’s an incomplete PR for this None .
I wouldn’t say it’s related RBAC because the issue seems Networking realted so connection timed out
ok so you are on chart major verison 4 while we are now on 6. Let me check one minute pls