Hello, I'm facing a terrible issue since I update the driver csi and kubernetes.
Everything is working with the drivers v3.0.0 but i can"t unmount persistent volumes when a pod is scale to 0.
I check the log and see that there is a difference between the commands pass by the driver to identificate a device and to remove it from the virtual machines.
Here is my log, from cluster A (old driver version but it works) and cluster (last driver version but not working).
Cluster A (working):
driver csi : v2.2.13
os : ubuntu 18.04
kubernetes version: 1.18.5
2022-08-19 10:37:10.886756 52046 [ERROR]: [requestID:3583979891] Can not get DMDevice by alias: dm-5
2022-08-19 10:37:10.886792 52046 [ERROR]: [requestID:3583979891] Get DMDevice by alias:dm-5 failed. error: Can not get DMDevice by alias: dm-5
2022-08-19 10:37:10.886816 52046 [ERROR]: [requestID:3583979891] check device: dm-5 is a partition device failed. error: Get DMDevice by alias:dm-5 failed. error: Can not get DMDevice by alias: dm-5
2022-08-19 10:37:10.886841 52046 [ERROR]: [requestID:3583979891] Get device of WWN 678b46a10025fc41736cbc0e0000007b error: check device: dm-5 is a partition device failed. error: Get DMDevice by alias:dm-5 failed. error: Can not get DMDevice by alias: dm-5
As you can see there is a difference in Gonna run shell cmd "ls -l /dev/mapper/ | grep -w dm-5". , in cluster A there is a awk and a print that I don't have in cluster B.
You can add the following parameters to the /etc/multipath.conf file by referring to section 3.6.1.2.
defaults {
user_friendly_names yes
find_multipaths no
}
Hello, I'm facing a terrible issue since I update the driver csi and kubernetes.
Everything is working with the drivers v3.0.0 but i can"t unmount persistent volumes when a pod is scale to 0.
I check the log and see that there is a difference between the commands pass by the driver to identificate a device and to remove it from the virtual machines.
Here is my log, from cluster A (old driver version but it works) and cluster (last driver version but not working).
Cluster A (working): driver csi : v2.2.13 os : ubuntu 18.04 kubernetes version: 1.18.5
Cluster B (not working): driver csi : v3.0.0 os : ubuntu 18.04 kubernetes version: 1.23.5
As you can see there is a difference in Gonna run shell cmd "ls -l /dev/mapper/ | grep -w dm-5". , in cluster A there is a awk and a print that I don't have in cluster B.
Here is how I construct my dockerfile :