Proxmox 4.1 - Error: /proc must be mounted

[hregis]

Hello,

I have a Debian Jessie with Proxmox 4.1, I have a dozen virtual servers in LXC, basically everything works. Randomly, all my VM lose access to process and this error message:

Error: /proc must be mounted
To mount /proc at boot you need an /etc/fstab line like:
proc /proc proc defaults
In the meantime, run "mount proc /proc -t proc"

I am forced to restart all VMs to resolve the issue.
By examining the logs of the host server, I find this :

Jan 07 06:25:06 ns338919 kernel: traps: lxcfs[2303] general protection ip:7fed0a7bb528 sp:7fed0953a730 error:0 in libc-2.19.so[7fed0a742000+19f000]
Jan 07 06:25:06 ns338919 systemd[1]: lxcfs.service: main process exited, code=killed, status=11/SEGV
Jan 07 06:25:07 ns338919 systemd[1]: Unit lxcfs.service entered failed state.
Jan 07 06:25:07 ns338919 systemd[1]: lxcfs.service holdoff time over, scheduling restart.
Jan 07 06:25:07 ns338919 systemd[1]: Stopping FUSE filesystem for LXC...
Jan 07 06:25:07 ns338919 systemd[1]: Starting FUSE filesystem for LXC...
Jan 07 06:25:07 ns338919 systemd[1]: Started FUSE filesystem for LXC.
Jan 07 06:25:07 ns338919 lxcfs[21524]: hierarchies: 1: name=systemd
Jan 07 06:25:07 ns338919 lxcfs[21524]: 2: cpuset
Jan 07 06:25:07 ns338919 lxcfs[21524]: 3: cpu,cpuacct
Jan 07 06:25:07 ns338919 lxcfs[21524]: 4: blkio
Jan 07 06:25:07 ns338919 lxcfs[21524]: 5: memory
Jan 07 06:25:07 ns338919 lxcfs[21524]: 6: devices
Jan 07 06:25:07 ns338919 lxcfs[21524]: 7: freezer
Jan 07 06:25:07 ns338919 lxcfs[21524]: 8: net_cls,net_prio
Jan 07 06:25:07 ns338919 lxcfs[21524]: 9: perf_event
Jan 07 06:25:07 ns338919 lxcfs[21524]: 10: hugetlb
Jan 07 06:25:07 ns338919 kernel: cgroup: new mount options do not match the existing superblock, will be ignored
Jan 07 06:25:07 ns338919 systemd[1]: Stopping PVE API Proxy Server...
Jan 07 06:25:17 ns338919 pveproxy[2050]: received signal TERM
Jan 07 06:25:17 ns338919 pveproxy[2050]: server closing
Jan 07 06:25:17 ns338919 pveproxy[2052]: worker exit
Jan 07 06:25:17 ns338919 pveproxy[2051]: worker exit
Jan 07 06:25:17 ns338919 pveproxy[2053]: worker exit
Jan 07 06:25:17 ns338919 pveproxy[2050]: worker 2051 finished
Jan 07 06:25:17 ns338919 pveproxy[2050]: worker 2053 finished
Jan 07 06:25:17 ns338919 pveproxy[2050]: worker 2052 finished
Jan 07 06:25:17 ns338919 pveproxy[2050]: server stopped
Jan 07 06:25:18 ns338919 systemd[1]: Starting PVE API Proxy Server...
Jan 07 06:25:19 ns338919 pveproxy[21788]: starting server
Jan 07 06:25:19 ns338919 pveproxy[21788]: starting 3 worker(s)
Jan 07 06:25:19 ns338919 pveproxy[21788]: worker 21789 started
Jan 07 06:25:19 ns338919 pveproxy[21788]: worker 21790 started
Jan 07 06:25:19 ns338919 pveproxy[21788]: worker 21791 started
Jan 07 06:25:19 ns338919 systemd[1]: Started PVE API Proxy Server.
Jan 07 06:25:19 ns338919 systemd[1]: Stopping PVE SPICE Proxy Server...
Jan 07 06:25:19 ns338919 spiceproxy[2059]: received signal TERM
Jan 07 06:25:19 ns338919 spiceproxy[2059]: server closing
Jan 07 06:25:19 ns338919 spiceproxy[2060]: worker exit
Jan 07 06:25:19 ns338919 spiceproxy[2059]: worker 2060 finished
Jan 07 06:25:19 ns338919 spiceproxy[2059]: server stopped
Jan 07 06:25:20 ns338919 systemd[1]: Starting PVE SPICE Proxy Server...
Jan 07 06:25:21 ns338919 spiceproxy[21811]: starting server
Jan 07 06:25:21 ns338919 spiceproxy[21811]: starting 1 worker(s)
Jan 07 06:25:21 ns338919 spiceproxy[21811]: worker 21812 started
Jan 07 06:25:21 ns338919 systemd[1]: Started PVE SPICE Proxy Server.
Jan 07 06:25:22 ns338919 systemd[1]: Stopping Proxmox VE firewall logger...
Jan 07 06:25:22 ns338919 pvepw-logger[800]: received terminate request (signal)
Jan 07 06:25:22 ns338919 pvepw-logger[800]: stopping pvefw logger
Jan 07 06:25:22 ns338919 systemd[1]: Starting Proxmox VE firewall logger...
Jan 07 06:25:22 ns338919 pvepw-logger[21842]: starting pvefw logger
Jan 07 06:25:22 ns338919 systemd[1]: Started Proxmox VE firewall logger.

My system reports is in attach-files

Thank you for your help

Régis

 

[windinternet]

You are not the only one. Problem is still being hunted down. LXCfs is necessary for the containers to see their uptime and memory usage combined with the ability to run systemd.

https://forum.proxmox.com/threads/cannot-open-proc-stat-transport-endpoint-is-not-connected.25225/