[
https://issues.apache.org/jira/browse/HDDS-13397?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18001383#comment-18001383
]
Attila Doroszlai edited comment on HDDS-13397 at 7/7/25 6:04 AM:
-----------------------------------------------------------------
I couldn't reproduce it with k3s. Can you please share logs from the failing
pods?
Based on code=403 for s3g, maybe it only happens with security enabled?
was (Author: adoroszlai):
I couldn't reproduce it with k3s.
> S3 and HttpFS gateways are failing on deployment from Kubernetes resource
> examples
> ----------------------------------------------------------------------------------
>
> Key: HDDS-13397
> URL: https://issues.apache.org/jira/browse/HDDS-13397
> Project: Apache Ozone
> Issue Type: Bug
> Components: HttpFS GateWay, kubernetes, s3gateway
> Affects Versions: 2.0.0
> Reporter: Denis Krivenko
> Priority: Major
>
> Ozone 2.0 cannot be successfully deployed to Kubernetes using provided
> resource examples because of constantly failing liveness probes.
>
> How to reproduce:
> 1. Deploy Ozone 2.0 from `kubernetes/examples/minikube`
> {code:java}
> ozone-2.0.0 > cd kubernetes/examples/minikube
> minikube > kubectl apply -f .
> {code}
> 2. Check pod statuses
> {code:java}
> minikube > kubectl get pod
> NAME READY STATUS RESTARTS AGE
> datanode-0 1/1 Running 0 10m
> datanode-1 1/1 Running 0 10m
> datanode-2 1/1 Running 0 10m
> httpfs-0 0/1 CrashLoopBackOff 6 (2m25s ago) 10m
> om-0 1/1 Running 0 10m
> recon-0 1/1 Running 0 10m
> s3g-0 0/1 CrashLoopBackOff 6 (2m25s ago) 10m
> scm-0 1/1 Running 0 10m {code}
> 3. Check events
> {code:java}
> minikube > kubectl events
> ...
> 10m (x6 over 14m) Normal Created Pod/s3g-0 Created container: s3g
> 10m (x6 over 14m) Normal Started Pod/s3g-0 Started container s3g
> 10m (x6 over 14m) Normal Started Pod/httpfs-0 Started container
> httpfs
> 10m (x6 over 14m) Normal Created Pod/httpfs-0 Created container:
> httpfs
> 9m28s (x6 over 13m) Normal Killing Pod/s3g-0 Container s3g failed
> liveness probe, will be restarted
> 9m18s (x6 over 13m) Normal Killing Pod/httpfs-0 Container httpfs
> failed liveness probe, will be restarted
> 8m10s (x8 over 9m28s) Warning BackOff Pod/s3g-0 Back-off restarting
> failed container s3g in pod
> s3g-0_default(b047ddba-062e-4d02-b3e9-622222a5d1ba)
> 8m8s (x8 over 9m18s) Warning BackOff Pod/httpfs-0 Back-off restarting
> failed container httpfs in pod
> httpfs-0_default(1c2db5b1-5d02-46e7-90fd-841658a18c8d)
> 4m16s (x7 over 13m) Normal Pulled Pod/s3g-0 Container image
> "apache/ozone:2.0.0" already present on machine
> 4m11s (x7 over 13m) Normal Pulled Pod/httpfs-0 Container image
> "apache/ozone:2.0.0" already present on machine
> 3m38s (x22 over 13m) Warning Unhealthy Pod/s3g-0 Liveness probe failed:
> HTTP probe failed with statuscode: 403
> 3m37s (x18 over 13m) Warning Unhealthy Pod/httpfs-0 Liveness probe failed:
> HTTP probe failed with statuscode: 500
> {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
