Good day all,
Short description of the issue: inability to start containers consistently.
Context in which the problem arose: 7-node cluster recently upgraded from version 7 to 8 (I believe the issue did not occur before the upgrade); no PVE subscription.
Scope: All the containers (three in number, each of them having an RBD-based root FS and an RBD-based additional volume
After some tests I have come to describe the issue as my Proxmox VE 8 seemingly not mapping all RBD upon container start on my fifth node. Here are samples of the two most relevant messages I managed to obtain:
More details follow. Please note that in all the code blocks I have substituted
Thank you in advance for your help.
Short description of the issue: inability to start containers consistently.
Context in which the problem arose: 7-node cluster recently upgraded from version 7 to 8 (I believe the issue did not occur before the upgrade); no PVE subscription.
Scope: All the containers (three in number, each of them having an RBD-based root FS and an RBD-based additional volume
mp0
) present on the fifth node.After some tests I have come to describe the issue as my Proxmox VE 8 seemingly not mapping all RBD upon container start on my fifth node. Here are samples of the two most relevant messages I managed to obtain:
Code:
lxc pre-start produced output: /dev/rbd0
lxc pre-start produced output: failed to get device path
Code:
lxc pre-start produced output: /dev/rbd1
lxc pre-start produced output: mount: /var/lib/lxc/.pve-staged-mounts/mp0:
special device /dev/rbd-pve/${cluster_id}/VirtualDisks/vm-122-disk-1 does not exist.
More details follow. Please note that in all the code blocks I have substituted
${cluster_id}
for the literal cluster ID.Thank you in advance for your help.
Last edited: