Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Hello, For Deployment Of Clearml Server In K8S, We Need To Label Just One Node App=Clearml And Also Set The Daemon.Json. Wouldn'T This Make Clearml Server Tie To This Node Only And If This Node Fails, Clearml Server Will Not Work. Any Advice? Similarly

Hello, for deployment of clearml server in k8s, we need to label just one node app=clearml and also set the daemon.json. wouldn't this make clearml server tie to this node only and if this node fails, clearml server will not work. Any advice?

Similarly for the limits for elastic search, how should we manage this with k8s setup?

Posted 2 years ago
Votes Newest

Answers 5

Hi SuccessfulKoala55 Thanks for pointing me to this repo. Was using this repo.

I didn't manage to find in this repo that if we still require to label the node app=clearml, like what was mentioned in the deprecated repo. Although from the values.yaml, the node selector is empty. Would u be able to advise?

How is the clearml data handled now then? Thanks

Posted 2 years ago

Hi OddShrimp85 ,
The app=clearml label is not required any more (in fact, it was only required in the initial, deprecated repo, and was removed in all following versions).
The data is handled now using PVs configured for the ElasticSearch and MongoDB charts. JuicyFox94 - would you like to expand? 🙂

Posted 2 years ago

Sure, OddShrimp85 , until you need to specifically bind any pod to a node, nodeSelector is not needed. In fact, the new chart will leave to k8s the right to share the load on the worker nodes. About pvc you simply need to declare the Storageclass at k8s level so it can take care of creating the PVC too. How many workers do you have in your setup?

Posted 2 years ago

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

Posted 2 years ago

Cool thanks guys. I am clearer now. Was confused by the obsolete info. Thanks for the clarification.

Posted 2 years ago
5 Answers
2 years ago
10 months ago