I have a handful of turnkey Linux (Debian 7) containers which have been happily chugging along for months/years on my Proxmox culster. Our VM storage is currently handled by a dedicated Ubuntu 14.04 machine using NFS 3 with no authentication. Recently, in a moment of extreme stupidity, I ran an update on the storage server without first shutting down all the dependent VMs. Unfortunately, one of the packages updated was nfs-kernal-server and you can guess what happened when that connection went down briefly.
After rebooting my Proxmox cluster all the KVM machines booted up and were/are perfectly fine. None of my containers (many of which use the same NFS shares as some KVM VMs) would start. Here's what happens:
This now happens for any NFS backed container (even new ones) running any OS (even tried Redhat, GRML, etc for funzies). When those same containers are backed up (from and to NFS shares) and then started on local storage, everything works. This makes no sense to me so I'm hoping someone can get me started on sorting this all out.
After rebooting my Proxmox cluster all the KVM machines booted up and were/are perfectly fine. None of my containers (many of which use the same NFS shares as some KVM VMs) would start. Here's what happens:
- Hit start in Proxmox web interface/OpenVZ CLI
- Machine mounts it's storage and shows as started (looks totally normal)
- Ram usage goes up to about 8-9MB and stops
- No more CPU usage, and no errors in the CLI
- Container shows as running but fails to boot and hangs
This now happens for any NFS backed container (even new ones) running any OS (even tried Redhat, GRML, etc for funzies). When those same containers are backed up (from and to NFS shares) and then started on local storage, everything works. This makes no sense to me so I'm hoping someone can get me started on sorting this all out.