what is management LIF in ONTAP SAN BAckend config, is it netapp cluster management LIF or SVM's Management IP, As We can't give netapp mgmt becuse of security restrictions but putting svm mgmt is giving problem like no storage pool assigned, even though the volume exists for svm.
#what is management LIF in ONTAP SAN BAckend config, is it netapp cluster management LIF or SVM's IP
verbal berry
verbal berry
thankyou, by using SVM mgmt interface it was giving me error on no storagepool assigned, even though I have created LUNs and volumes for the svm. could it be any role issue as the user I used was the svm admin user vsadmin
opal steeple
you do not need to create LUNs for the SVM; Trident creates all the LUNs/Volumes as soon as a matching PVC comes in.
As to the error, without seeing the actual error message in context I cannot help you there (there are so many things named "pool" in this scenario that it's hard to know what your error refers to without seeing it)
verbal berry
Thankyou, yes without creating any LUNs or volumes directly and assigning local tier, I can see trident is able to create volumes on netapp. thankyou.
another problem is very specific to openshift, I can see pvc is bound and volume is there, but pod is not able to mount, saying failed to mount. even though the volume is attached to it.
opal steeple
yeah you probably need to install the iscsi tools on the worker nodes. the documentation describes this for a couple different distributions
verbal berry
yes thats there , also I am now trying with NAS
opal steeple
If it's not iscsi tools installation, then it could be CHAP or one of many other issues. more detailed logs from the worker node and the ONTAP system would be required to troubleshoot this
verbal berry
ok any specific logs on worker , because I can see on worker the volume is mounted as folder , but pod is not able to mount it
Normal SuccessfulAttachVolume 13m attachdetach-controller AttachVolume.Attach succeeded for volume "pvc-xxx"
Warning FailedMount 2m3s (x5 over 11m) kubelet Unable to attach or mount volumes: unmounted volumes=[my-volume], unattached volumes=[], failed to process volumes=[]: timed out waiting for the condition
Warning FailedMount 49s (x6 over 11m) kubelet MountVolume.SetUp failed for volume "pvc-xxxxxx" : rpc error: code = DeadlineExceeded desc = context deadline exceeded
this is my pod log
opal steeple
Doesn't seem to be mounting the volume. You say you see it mounted? Is it accesible from the worker? Can you write to it?
verbal berry
I think problem is it is not able to acquire lock on volume -
time="2024-12-24T02:25:54Z" level=debug msg="Attempting to acquire shared lock (NodePublishVolume-pvc-xxxx); 9 position in the queue." lock=csi_node_server logLayer=csi_frontend requestID=addd3674-0165-4f3b-8bb1-f1540ff15d0f requestSource=CSI workflow="node_server=publish"
Can't see the next log that it is acquired.
this is trident-node logs for specific worker
on worker I can see the folder, but I think its not mouted.
but volume on netapp is created
trying with mount.nfs command locally on woker, as seems its not working
verbal berry
local mount fs command on worker also doesn't work, I can see status, portmapper, nfslockmanager, mountd all processes runing on svm, and on local postmapper is running ( verified thourgh rpcinof), telnet on port 2049 /111 works from client to netapp nfs data LIf.
shwmount shows the volume path created by trident.
But mount command is still stuck and not retuning any output, gives timeut after long.
also I have added the subnet in export list of SVM with all users permisions.
