Search results

I'm not home right now but I've used proxmox before without this issue, even while installing on top of buster. I've verified this issue on multiple nodes and ifupdown2 is installed. Steps to reproduce: Install buster. Install proxmox on top. Install opensm/rdma-core and follow the how to for...

It seems to be throwing an error about the length of mac address (24) not being supported.

TASK ERROR: command 'ifreload -a' failed: exit code 1 How can I fix this? Happens every time I modify the network.

I was able to replicate the issue on other distros so it must be something with my configuration. I'm likely not understanding how to setup rdma in the configuration file or my switch doesn't support it. It's a fairly cheap switch and it's unmanaged. Either way ipoib works for me. I'll keep...

When using rdma with connect-x 3 cards ceph crashes. I know rdma support is iffy with ceph and don't mind sticking with ipoib but it would be nice if it did work.

I want to move the public/private network of ceph over to infiniband which is on a different subnet. Is there an easy way to do this?

The documents say you cant change the hostname if the server is in a cluster. Ceph does not like my hostnames (MDS): " $ MDS id 'mds.01-8700k' is invalid. MDS names may not start with a numeric digit" Do I actually need to reinstall my entire cluster? Or can this be salvaged.

After having just experienced this please have proxmox backup all relevant OSD/Ceph data so an encrypted array can be recovered. I'm aware I should have made backups of the configuration (I did have recent file backups). However, corosync shouldn't be able to just delete those files on every...

While removing a node from cluster and adding a new one I stupidly created a new ceph which somehow affected all nodes. rbd and ceph.conf were wiped out in /etc/ceph. I have the FSID. There are encrypted OSD (OSD are intact). /var/lib/ceph is intact but a lot of it was overwritten. I have the...