k8s-csi-s3 icon indicating copy to clipboard operation
k8s-csi-s3 copied to clipboard

I applied pod.yaml file but pod is in container-creating state

Open Venkatesh7591 opened this issue 1 year ago • 4 comments

Getting error in the pod is still container creating state " Events: Type Reason Age From Message


Normal Scheduled 42s default-scheduler Successfully assigned default/csi-s3-test-nginx to 10.0.10.21 Warning FailedMount 10s (x7 over 42s) kubelet MountVolume.MountDevice failed for volume "pvc-f92e57af-8a49-4b4b-8910-c301013e0767" : rpc error: code = Unknown desc = Error starting systemd unit geesefs-pvc_2df92e57af_2d8a49_2d4b4b_2d8910_2dc301013e0767.service on host: Cannot set property ExecStopPost, or unknown property.

"

Venkatesh7591 avatar Sep 08 '23 12:09 Venkatesh7591

I changed to mounter-s3fs also but same issue " Events: Type Reason Age From Message


Normal Scheduled 59s default-scheduler Successfully assigned default/csi-s3-test-nginx to 10.0.10.21 Warning FailedMount 27s (x7 over 59s) kubelet MountVolume.MountDevice failed for volume "pvc-04fa6a57-e491-4c83-8a57-631b9081e1c6" : rpc error: code = Unknown desc = Error fuseMount command: s3fs args: [face-biometrics:/pvc-04fa6a57-e491-4c83-8a57-631b9081e1c6 /var/lib/kubelet/plugins/kubernetes.io/csi/ru.yandex.s3.csi/f34fb1217de73f2aff6c6ecd73e03893a8a5e52aa61896f3cad9a9e8f8cc10ea/globalmount -o use_path_request_style -o url=https://idq7snkhlnmv.compat.objectstorage.us-ashburn-1.oraclecloud.com -o allow_other -o mp_umask=000 -o endpoint=us-ashburn-1 --memory-limit 1000 --dir-mode 0777 --file-mode 0666] output:

"

Venkatesh7591 avatar Sep 08 '23 12:09 Venkatesh7591

I have the same issue

image

zhaoxin-BF avatar Nov 30 '23 09:11 zhaoxin-BF

Cannot set property ExecStopPost means that your systemd is too old. csi-s3 uses a tricky way to start geesefs on host so it doesn't break when you restart the csi-s3 pod. Either upgrade systemd or add --no-systemd mount option.

vitalif avatar Dec 27 '23 09:12 vitalif

Cannot set property ExecStopPost means that your systemd is too old. csi-s3 uses a tricky way to start geesefs on host so it doesn't break when you restart the csi-s3 pod. Either upgrade systemd or add --no-systemd mount option.

set options as below, pod will running successfully.

image

tomcheung0 avatar Jan 26 '24 07:01 tomcheung0