The clearML agent works within the Kubernetes cluster
Hi @<1669152726245707776:profile|ManiacalParrot65> , is this a specific task or the controller?
Its actually happens in both sometimes in the pipeline task and sometimes in pipeline controller
I just found out that ClearML Agent has a service mode. However, I'm currently using ClearML Agent with a Helm chart on Kubernetes (K8s). How can I start the agent in service mode in this setup?
@<1523701070390366208:profile|CostlyOstrich36> can you help me with that? I can provide you more information if you need 🙌
@<1669152726245707776:profile|ManiacalParrot65> could you please send your values file override for the Agent helm chart?
@<1729671499981262848:profile|CooperativeKitten94> here:
None
global:
imageRegistry: "docker-proxy.nexmart.com:5000"
clearml:
existingAgentk8sglueSecret: "clearml-agent-secret"
existingClearmlConfigSecret: "clearml-agent-secret"
agentk8sglue:
defaultContainerImage: "repo.nexmart.com:5000/nm-container-python:3.9"
apiServerUrlReference: " None "
fileServerUrlReference: " None "
webServerUrlReference: " None "
Use SA from default agent
serviceExistingAccountName: "test-clearml-clearml-agent-sa"
queue: "services"
initContainers:
resources:
requests:
memory: "50M"
cpu: "50m"
limits:
memory: "200M"
cpu: "1"
resources:
requests:
memory: "50M"
cpu: "10m"
limits:
memory: "500M"
cpu: "500m"
basePodTemplate:
env:
- name: "GL_PACA_PIPELINE_TRIGGER_TOKEN"
valueFrom:
secretKeyRef:
key: "gl_paca_trigger_token"
name: "clearml-agent-mlops-secret"
resources:
requests:
memory: "50M"
cpu: "10m"
limits:
memory: "750M"
cpu: "500m"
@<1729671499981262848:profile|CooperativeKitten94> I still have this problem even with other pipelines, can you please help me?