https://flyte.org
Join Slack
Does the sandbox cluster support running in a podman rootless cluster?
# flyte-support
l
Does the sandbox cluster support running in a podman rootless cluster?
f
Hi Endre we have not tried that
Please let us know, you can use the helm chart direct
We want to have a simplified startin experience hence we package Flyte into K3d like environment- using flytectl, but we are just deploying a helm chart
l
@freezing-airport-6809 I tried but stumped upon a problem about it complaining about no root access, let me get you logs
f
Hmm, you should use The newer version of all components. These should not need root access
Is this the user container
l
podman logs flyte-sandbox 5m 36s task 4 180526 Starting Docker daemon... Done. Starting k3s cluster... Done. Deploying Flyte... Error: open /root/.config/helm/repositories.lock: permission denied
f
cc @great-school-54368?
g
@limited-hairdresser-5419 Can you send me what command you are running for sandbox ? Are you using `--source`I flag in 
flytectl sandbox start
command ? If yes then i found some workaround for podman. Please let us know if you are facing issue with simple 
flytectl sandbox start
It's docker command for running sandbox 
docker run -it  --name flyte-sandbox --privileged -p 30081:30081 <http://ghcr.io/flyteorg/flyte-sandbox|ghcr.io/flyteorg/flyte-sandbox>
l
flytectl sandbox start --source .
g
does 
flytectl sandbox start
is working without 
--source
flag
l
Nope 😢
Copy code
/ # k3s kubectl get pod 
No resources found in default namespace.
/ # k3s kubectl get pod  -A
NAMESPACE     NAME                                  READY   STATUS    RESTARTS   AGE
flyte         flyte-contour-contour-certgen-k8f8n   0/1     Pending   0          3m26s
kube-system   coredns-7448499f4d-z897j              0/1     Pending   0          3m26s
/ # k3s kubectl describe pod -n kube-system 
Name:                 coredns-7448499f4d-z897j
Namespace:            kube-system
Priority:             2000000000
Priority Class Name:  system-cluster-critical
Node:                 <none>
Labels:               k8s-app=kube-dns
                      pod-template-hash=7448499f4d
Annotations:          <none>
Status:               Pending
IP:                   
IPs:                  <none>
Controlled By:        ReplicaSet/coredns-7448499f4d
Containers:
  coredns:
    Image:       rancher/coredns-coredns:1.8.3
    Ports:       53/UDP, 53/TCP, 9153/TCP
    Host Ports:  0/UDP, 0/TCP, 0/TCP
    Args:
      -conf
      /etc/coredns/Corefile
    Limits:
      memory:  170Mi
    Requests:
      cpu:        100m
      memory:     70Mi
    Liveness:     http-get http://:8080/health delay=60s timeout=1s period=10s #success=1 #failure=3
    Readiness:    http-get http://:8181/ready delay=0s timeout=1s period=2s #success=1 #failure=3
    Environment:  <none>
    Mounts:
      /etc/coredns from config-volume (ro)
      /var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-jgl97 (ro)
Conditions:
  Type           Status
  PodScheduled   False 
Volumes:
  config-volume:
    Type:      ConfigMap (a volume populated by a ConfigMap)
    Name:      coredns
    Optional:  false
  kube-api-access-jgl97:
    Type:                    Projected (a volume that contains injected data from multiple sources)
    TokenExpirationSeconds:  3607
    ConfigMapName:           kube-root-ca.crt
    ConfigMapOptional:       <nil>
    DownwardAPI:             true
QoS Class:                   Burstable
Node-Selectors:              <http://beta.kubernetes.io/os=linux|beta.kubernetes.io/os=linux>
Tolerations:                 CriticalAddonsOnly op=Exists
                             <http://node-role.kubernetes.io/control-plane:NoSchedule|node-role.kubernetes.io/control-plane:NoSchedule> op=Exists
                             <http://node-role.kubernetes.io/master:NoSchedule|node-role.kubernetes.io/master:NoSchedule> op=Exists
                             <http://node.kubernetes.io/not-ready:NoExecute|node.kubernetes.io/not-ready:NoExecute> op=Exists for 300s
                             <http://node.kubernetes.io/unreachable:NoExecute|node.kubernetes.io/unreachable:NoExecute> op=Exists for 300s
Events:
  Type     Reason            Age    From               Message
  ----     ------            ----   ----               -------
  Warning  FailedScheduling  3m42s  default-scheduler  no nodes available to schedule pods
  Warning  FailedScheduling  3m41s  default-scheduler  no nodes available to schedule pods
@great-school-54368 any idea ? ^
g
@limited-hairdresser-5419 sorry i didn't get time to debug the issue, I think it's a bug. We never tested the setup with podman. Please create a issue with as much as information possible and tag me. Having conversation on issue will help others,
f
.. (edited)
cc @high-park-82026 we could probably use podman to run all flyte backend components as a pod?
288 Views
Open in Slack
PreviousNext
Powered by

Flyte

Flyte enables production-grade orchestration for machine learning workflows and data processing created to accelerate local workflows to production.

Powered by