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
Answered
Hi Folks, I Am Trying To Run Clearml On A Local Kubernetes (Created With Rancher Desktop), But I Am Running Into Issues. I Admit I Am A Complete Noob To K8S, Helm, Clearml Etc. Etc. So I Am Not Sure If I Did Some Very Stupid Mistakes. I Have A K8S Cluste

Hi folks, I am trying to run ClearML on a local kubernetes (created with Rancher Desktop), but I am running into issues.
I admit I am a complete noob to K8s, helm, ClearML etc. etc. so I am not sure if I did some very stupid mistakes.

I have a K8s cluster running managed by Rancher Desktop.
Next, I went to https://github.com/allegroai/clearml-helm-charts and following the instructions I typed the following in my terminal:

helm repo add allegroai helm repo update helm install clearml allegroai/clearml
This last command shows the following:

` W0502 16:30:54.761799 16570 warnings.go:70] policy/v1beta1 PodDisruptionBudget is deprecated in v1.21+, unavailable in v1.25+; use policy/v1 PodDisruptionBudget
W0502 16:30:54.826250 16570 warnings.go:70] policy/v1beta1 PodDisruptionBudget is deprecated in v1.21+, unavailable in v1.25+; use policy/v1 PodDisruptionBudget
NAME: clearml
LAST DEPLOYED: Mon May 2 16:30:54 2022
NAMESPACE: default
STATUS: deployed
REVISION: 1
NOTES:

  1. Get the application URL by running these commands:
    export NODE_PORT=$(kubectl get --namespace default -o jsonpath="{.spec.ports[0].nodePort}" services clearml)
    export NODE_IP=$(kubectl get nodes --namespace default -o jsonpath="{.items[0].status.addresses[0].address}")
    echo http://$NODE_IP:$NODE_PORT I followed the advice in the notes, but it looks like something is wrong, when I type: export NODE_PORT=$(kubectl get --namespace default -o jsonpath="{.spec.ports[0].nodePort}" services clearml)
    export NODE_IP=$(kubectl get nodes --namespace default -o jsonpath="{.items[0].status.addresses[0].address}")
    echo http://$NODE_IP:$NODE_PORT it looks like the "clearml" service doesn't exist and NODE_PORT is empty: Error from server (NotFound): services "clearml" not found
    : `
    Do you have any advice to debug what's going on and start ClearML on my local kubernetes cluster?
  
  
Posted 2 years ago
Votes Newest

Answers 31


many thanks πŸ™‚ I am going to play with ClearML a little bit and re-read carefully the thread to learn something from what you made me do today!

  
  
Posted 2 years ago

I’m going to release a patch very soon

  
  
Posted 2 years ago

you will need to upgrade clearml helm chart

  
  
Posted 2 years ago

yes

  
  
Posted 2 years ago

3.10.2 will be published in 30 minutes

  
  
Posted 2 years ago

so you installed chart version 3.10.1

  
  
Posted 2 years ago

yw

  
  
Posted 2 years ago

this is the PR https://github.com/allegroai/clearml-helm-charts/pull/66 ; when CI will be passed I’m going to release the new chart

  
  
Posted 2 years ago

so we also improved the chart, superhappy πŸ˜„

  
  
Posted 2 years ago

Chart 3.10.2 released

  
  
Posted 2 years ago

It did work Valeriano!

  
  
Posted 2 years ago

oh yass

  
  
Posted 2 years ago

just to understand well the problems you helped me fix:
for elastic search it looked like I wasn't running the cluster with enough memory

but what happened to the FileServer? and how can I prevent it happening in a potential "production" deployment?

  
  
Posted 2 years ago

for fileserver the persistent volume need to be provisioned by a storageclass. Ususally I always set it to standard becase it’s commonly used in public cloud providers

  
  
Posted 2 years ago

i will release a new chart version with no need to set a default storage class like I asked you to do today

  
  
Posted 2 years ago

yay πŸ™‚

  
  
Posted 2 years ago

wohoo

  
  
Posted 2 years ago

Thanks a lot!

  
  
Posted 2 years ago

but I’m going to patch this soon so it will take the default storageclass automatically

  
  
Posted 2 years ago

now I’m leaving for today but let me know if it works. Have an nice day!

  
  
Posted 2 years ago

with this one you don’t need anymore to set the value I suggested before

  
  
Posted 2 years ago

you should be able to do a helm repo update

  
  
Posted 2 years ago

I absolutely need to improve the persistence part of this chart πŸ˜„

  
  
Posted 2 years ago

thanks a lot! So as long as we have the storageclass in our kubernetes cluster configured correctly, the new helm chart should work out of the box?

  
  
Posted 2 years ago

perfect, let me try πŸ™‚ (thanks a lot for all the help!)

  
  
Posted 2 years ago

thanks a lot πŸ™‚ that was quick πŸ™‚

  
  
Posted 2 years ago

using the --set you adviced above right?

  
  
Posted 2 years ago

I;m already testing this patch

  
  
Posted 2 years ago

yes

  
  
Posted 2 years ago

and see the new version

  
  
Posted 2 years ago
22K Views
31 Answers
2 years ago
8 months ago
Tags
Similar posts