Thanks for the report 🙂 - we'll put it in the pipeline 🙂

Hi RobustRat47 , can't personal access tokens be used as passwords?

Hi JuicyFox94 ,
Thanks! 🙏
I'll ask someone to take a look as soon as possible - will keep you updated 🙂

Hi OddShrimp85 , I'll take a look. It basically means the glue failed to push the task to the intermediate queue. Can you please open a GitHub issue?
Also, is the task spinning to and running, or does this error prevents you from using it?

Hi NarrowLion8 , this is currently not recorded in the system

Hi @<1558262042589138944:profile|FantasticPig28> , check out clearml-task (its a CLI automatically installed as part of the clearml SDK)

WickedCat12 we've identified the problem and will push a fix on the next update, thanks for reporting!

WickedCat12 a fix has been pushed - please see if it works for you 🙂

Again a DNS issue with the .ml domains... Mostly with Google DNS Servers, it seems

WittyOwl57 can you show a screen shot of the experiments and their comparison?

It's still worth mentioning that HTTPS termination is up to the deployment, and is highly recommended if you deploy your server in an open network.

Hi WittyOwl57 , thanks for the great ideas - would you mind adding these to the UI suggestions thread in GitHub? https://github.com/allegroai/clearml/issues/81

Sure 🙂

SmugSnake85 clearml-data only supports local files which it manages and uploads to a remote storage location. Managing references to remote data (as it seems you're looking for) is something supported in the clearml paid version (called hyper datasets, you can read about it here - https://clear.ml/docs/latest/docs/hyperdatasets/overview )

From your description, it clearly seems that the Agents are connecting to a different Trains Server... This was the whole point of using a different queue name, since every server srats out with the default queue 🙂

I think there was some issue with MongoDB Atlas (someone mentioned it, not sure if the issue was resolved)

Can I see the complete docker-compose.yml you're using?

OK, good to know 😄

Hi ThickCrow45 ,
Adding HTTPS support is something that is usually done in an external layer (as you mentioned, a load balancer).
To do it, you can use NGINX (which is widely used) and configure it to act as a load-balancer with SSL termination. See for example:
https://docs.nginx.com/nginx/admin-guide/security-controls/securing-http-traffic-upstream/
https://www.digitalocean.com/community/tutorials/how-to-set-up-nginx-load-balancing-with-ssl-termination

In our case, the ClearML server ...

You can even add NGINX in the docker-compose file (or docker-compose override file) using the publicly-available NGINX docker containers (see https://hub.docker.com/_/nginx ) - this way you won't have to install anything, just properly configure the NGINX container 🙂

Will do 🙂

The best wait is to open an issue in GitHub ( https://github.com/allegroai/clearml-server )

Hi @<1529271085315395584:profile|AmusedCat74> , as far as I know we do not use the Proxy-Authorization header at all. Having said that, of course we'll check it out and do some testing. If you have some results testing with this new version, please do share 🙂

Hi @<1595587997728772096:profile|MuddyRobin9> , what clearml SDK and agent versions are you using? Also, can you share the complete task log?

Hi OddShrimp85 , from your question I see you've been looking at the old, deprecated helm chart. Please see the new helm chart at https://github.com/allegroai/clearml-helm-charts

Hi JollyChimpanzee19 , what's the exact use-case?

Hi CourageousWhale20 ,
While this is not part of the Task interface (perhaps it should be - we'll appreciate a GitHub issue on the subject), you can access this value as follows:
` task = Task.init(...)

branch = task.data.script.branch `

Is it mandatory, for example, to label the node app=clearml

It's only required in the legacy helm chart. If you're using the cloud-ready helm chart, or the even newer helm chart ( https://github.com/allegroai/clearml-helm-charts thanks JuicyFox94 !!) - that's not required at all.

Hi PleasantGiraffe85 - welcome to ClearML 😄
The ElasticSearch is used to index various aspects of the experiment and devops process, e.g. index experiment-specific metrics and events, execution agents performance reports etc.

Actually, if you use our helm charts, the ES part itself is not that complicated (since it uses the official ES chart), but I get why using k8s can be a bit more than simply using an existing ES installation 🙂