Bug 2211568 - Image-upload hangs when access mode is not specified
Summary: Image-upload hangs when access mode is not specified
Keywords:
Status: CLOSED INSUFFICIENT_DATA
Alias: None
Product: Container Native Virtualization (CNV)
Classification: Red Hat
Component: Storage
Version: 4.12.3
Hardware: Unspecified
OS: Unspecified
unspecified
low
Target Milestone: ---
: ---
Assignee: Adam Litke
QA Contact: Natalie Gavrielov
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2023-06-01 05:32 UTC by David Sedgmen
Modified: 2023-10-20 04:25 UTC (History)
1 user (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2023-06-21 12:49:41 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Issue Tracker CNV-29316 0 None None None 2023-06-01 05:39:51 UTC

Description David Sedgmen 2023-06-01 05:32:37 UTC
Description of problem:
Image-upload hangs when access mode is not specified because it will not create the pvc

Version-Release number of selected component (if applicable):

virtctl version
Client Version: version.Info{GitVersion:"v0.58.1-57-gfa16ad5c8", GitCommit:"fa16ad5c8189c14e3adf4e757dbfbc68216af85f", GitTreeState:"clean", BuildDate:"2023-05-11T05:38:34Z", GoVersion:"go1.18.10", Compiler:"gc", Platform:"linux/amd64"}
Server Version: version.Info{GitVersion:"v0.58.1-57-gfa16ad5c8", GitCommit:"fa16ad5c8189c14e3adf4e757dbfbc68216af85f", GitTreeState:"clean", BuildDate:"2023-05-11T05:40:21Z", GoVersion:"go1.19.6", Compiler:"gc", Platform:"linux/amd64"}

kubevirt-virtctl-4.12.3-1137.el8.x86_64
OpenShift version: 4.12.11

How reproducible:

Everytime

Steps to Reproduce:
1. Setup either iscsi persistent storage or nfs dynamic provisioner 
2. Try to create image dv 

Actual results:

~~~
[kni@localhost iscsi]$ virtctl image-upload dv openstack-base-img3 -n openstack --size 50Gi --image-path=/var/lib/libvirt/images/rhel-8.4-x86_64-kvm.qcow2 --insecure --storage-class=iscsi-ssd --block-volume
PVC openstack/openstack-base-img3 not found 
DataVolume openstack/openstack-base-img3 created
Waiting for PVC openstack-base-img3 upload pod to be ready...
^Z
[1]+  Stopped                 virtctl image-upload dv openstack-base-img3 -n openstack --size 50Gi --image-path=/var/lib/libvirt/images/rhel-8.4-x86_64-kvm.qcow2 --insecure --storage-class=iscsi-ssd --block-volume
[kni@localhost iscsi]$ bg
[1]+ virtctl image-upload dv openstack-base-img3 -n openstack --size 50Gi --image-path=/var/lib/libvirt/images/rhel-8.4-x86_64-kvm.qcow2 --insecure --storage-class=iscsi-ssd --block-volume &

[kni@localhost iscsi]$ oc get pvc
NAME                          STATUS   VOLUME                                     CAPACITY      ACCESS MODES   STORAGECLASS       AGE
openstack-base-img            Bound    pvc-22c71b48-e6a9-47d0-8547-31e882791889   56811736720   RWX            nfs-client-share   5h9m
openstack-base-img2           Bound    iscsi-pv-04                                100Gi         RWO            iscsi-ssd          4h58m <---- one created with  --access-mode ReadWriteOnce
openstackclient-cloud-admin   Bound    pvc-95ad1133-63ef-4ddc-8d90-0fbaf9f56a02   4G            RWO            nfs-client         26h
openstackclient-hosts         Bound    pvc-f49066ac-4824-46c1-81e4-ba181ec17ac3   1G            RWO            nfs-client         26h
openstackclient-kolla-src     Bound    pvc-894cf9ac-8b1b-4816-947f-cb2026e1b410   1G            RWO            nfs-client         26h
~~~

Expected results:

To use a default access-mode or to error out with a meaningful error 

Additional info:

Seems to be a reversion of https://bugzilla.redhat.com/show_bug.cgi?id=2126104

Comment 1 Yan Du 2023-06-14 12:18:37 UTC
It looks like caused by incompleted storage profile, could you please attach the storage class yaml and storage profile yaml? The storage profile has the same name as storage class.

Comment 2 David Sedgmen 2023-06-14 22:14:11 UTC
My Test environment has expired, so I not able to retrieve the SC any more. 

I replicated on 2 types storage profiles. 

One was using an upstream nfs provisoner https://github.com/kubernetes-sigs/nfs-subdir-external-provisioner/blob/master/charts/nfs-subdir-external-provisioner/README.md
Which generates a default SC and is meant to default to ReadWriteOnce. 

The other was iSCSI using the kubernetes.io/no-provisioner which I did try setting the access mode in, but it just dropped it.


kind: PersistentVolume
apiVersion: v1
metadata:
  name: iscsi-pv-01
  uid: 548684e5-36e1-4686-90db-bfe5b5b7c04b
  resourceVersion: '41320730'
  creationTimestamp: '2023-06-01T00:26:44Z'
  finalizers:
    - kubernetes.io/pv-protection
  managedFields:
    - manager: kube-controller-manager
      operation: Update
      apiVersion: v1
      time: '2023-06-01T00:26:44Z'
      fieldsType: FieldsV1
      fieldsV1:
        'f:status':
          'f:phase': {}
      subresource: status
    - manager: kubectl-create
      operation: Update
      apiVersion: v1
      time: '2023-06-01T00:26:44Z'
      fieldsType: FieldsV1
      fieldsV1:
        'f:spec':
          'f:accessModes': {}
          'f:capacity':
            .: {}
            'f:storage': {}
          'f:iscsi':
            .: {}
            'f:fsType': {}
            'f:iqn': {}
            'f:iscsiInterface': {}
            'f:lun': {}
            'f:targetPortal': {}
          'f:persistentVolumeReclaimPolicy': {}
          'f:storageClassName': {}
          'f:volumeMode': {}
spec:
  capacity:
    storage: 100Gi
  iscsi:
    targetPortal: '*******'
    iqn: '*******'
    lun: 0
    iscsiInterface: default
    fsType: ext4
  accessModes:
    - ReadWriteOnce
  persistentVolumeReclaimPolicy: Retain
  storageClassName: iscsi-ssd
  volumeMode: Filesystem
status:
  phase: Available

Comment 3 Yan Du 2023-06-21 12:49:41 UTC
Hi, David
Thanks for reporting the bug,
but unfortunately we still need the storage class and storage profile yaml for future debugging.
Please feel free to reopen it once you could get the storage class and storage profile information.

Comment 4 Yan Du 2023-06-21 12:49:54 UTC
*** Bug 2211571 has been marked as a duplicate of this bug. ***

Comment 5 Yan Du 2023-06-21 12:54:23 UTC
btw, could you please try the nfs with ReadWriteMany to see whether it is working?

Comment 6 Red Hat Bugzilla 2023-10-20 04:25:07 UTC
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 120 days


Note You need to log in before you can comment on or make changes to this bug.