Cleanup with CSI Driver for Dell EMC PowerMax

Question

Hello, I have a question with stoptest.sh on repository.   CSI Driver for Dell EMC PowerMax 1.2 CentOS 7.6 with Native Multipath,   1) Run deployment script (starttest.sh in the CSI guide) and confirm it creates SG/Vol/MV and paths on host. 2) Run remove script (stoptest.sh in the CSI guide), and SG/Vol/MV are removed.in host side and multipath disappear as well. However, 'lsblk' outputs some devices like sdb, sdc, etc but 'multipath -ll' outputs nothing. Is it as expected or should we run extra cleanup commands? Thank you, Satoshi

bmcfeeters · Answer

Hello Satoshi, You may need to run something like the rescan-scsi-bus.sh script from the sg3_utils package to remove the stale devices. ThanksBryan

Wellin · Answer

We have performed the rescan,but the path devices still there.

Wellin · Answer

Hello Sir.

Let me clarify the hole progress.

1. This situation not happened frequently .

2. During the CSI testing , i need to deploy/remove the pods of k8s,and the CSI driver in host side will rescan/release the multipath as well.

3. When i run the deployment script( starttest.sh in the CSI guide) ,new SG/Vol/MV will be created accordingly .in host side ,new paths will appear.

4. When i run the remove script( stoptest.sh in the CSI guide),SG/Vol/MV will be removed . In host side ,the multipath will disappear as well.

5. All of the above steps is fine,except the paths still there,only have devices,"lsblk" will output some device like sdb,sdc...,but "multipath -ll " output nothing,.--maybe need more commands to cleanup.

6. When i run the deployment script again, new SG/Vol/MV will be created normally,but in host side,the path DID NOT appear.

7.In this situation i have to perform a rescan,meanwhile ,CSI driver should running rescan as well,then below situation appears.

#####################################

t2-pmax-db010:~$ sudo multipath -ll
mpathe (360000970000497900099533030303342) dm-4 EMC ,SYMMETRIX
size=100G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=1 status=active
|- 13:0:4:1 sdc 8:32 active ready running
|- 14:0:2:1 sdi 8:128 active ready running
|- 13:0:5:1 sdf 8:80 active ready running
|- 14:0:2:2 sdj 8:144 active ready running
|- 13:0:5:2 sdg 8:96 active ready running
|- 14:0:3:1 sdl 8:176 active ready running
`- 14:0:3:2 sdm 8:192 active ready running
mpathd (360000970000497900099533030303433) dm-3 EMC ,SYMMETRIX
size=100G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=1 status=active
`- 13:0:4:2 sdd 8:48 active ready running
mpathc (360000970000497900099533030303341) dm-2 EMC ,SYMMETRIX
size=200G features='1 queue_if_no_path' hwhandler='0' wp=rw
`-+- policy='round-robin 0' prio=1 status=active
|- 13:0:4:0 sdb 8:16 active ready running
|- 14:0:2:0 sdh 8:112 active ready running
|- 13:0:5:0 sde 8:64 active ready running
`- 14:0:3:0 sdk 8:160 active ready running

#####################################

8.If i restart the host ,the issue disappeared.

9.It is not happened frequently .And i think it should be related to not cleaning up the multipath.

Appreciate for your replying.

bmcfeeters · Answer

Hi Wellin,

I think to get to the bottom of this one we will have to take a look at the logs during volume creation and deletion.

Please contact me via private message and we can get the troubleshooting going.

Thanks
Bryan

Containers

Was this post helpful?