Open marcinpro1 opened 4 days ago
hmm that error message is not very helpful tbh
@dsharma-dc the page-size we get from the blockbstore lvols is 4k right? Not sure how we'd get different block-size on the replicas...
@marcinpro1, could you please run this command on each node's backend device, example:
sudo blockdev --report /dev/nvme0n1
nvme-1:
[root@nvme-1 ~]# blockdev --report /dev/nvme2n1
RO RA SSZ BSZ StartSec Size Device
rw 8192 512 4096 0 1920383410176 /dev/nvme2n1
nvme-2:
[root@nvme-2 ~]# blockdev --report /dev/nvme3n1
RO RA SSZ BSZ StartSec Size Device
rw 8192 512 4096 0 1920383410176 /dev/nvme3n1
nvme-3:
[root@nvme-3 ~]# blockdev --report /dev/nvme2n1
RO RA SSZ BSZ StartSec Size Device
rw 8192 512 4096 0 1920383410176 /dev/nvme2n1
@tiagolobocastro The block size here is taken what's set in the replica namespace apparently. So perhaps replica namespace hasn't got populated correctly here. Could you provide the complete logs, primarily io-engine logs? Is the issue still persisting or was it transient?
Issue still persists. Here you go: openebs_dump.tar.gz
Cannot attach created pvc volume to pod.
Storage made on 3 nodes with 1 nvme (same model, same vendor) disk/diskpool per node
Logs from agent core:
Describe from pod: