Can't ping LXC after rename

[Leon Roy]

After changing an LXC container Hostname under its DNS setting, I can ping it, but can no longer SSH to it.

Other containers are accessible and the renamed container was accessible before the Hostname change. Is this a known issue or any suggestions?

Thanks.

 

[Leon Roy]

After running `lxc-attach ID` I ran the command:
~# /etc/init.d/ssh restart

[....] Restarting ssh (via systemctl): ssh.serviceFailed to add /run/systemd/ask-password to directory watch: No space left on device

Any ideas why it'd say 'No space left on device'?

df returns:

Filesystem 1K-blocks Used Available Use% Mounted on
volume1/virtualmachines/subvol-101-disk-0 8388608 541952 7846656 7% /
none 492 0 492 0% /dev
tmpfs 16456152 0 16456152 0% /dev/shm
tmpfs 16456152 8260 16447892 1% /run
tmpfs 5120 0 5120 0% /run/lock
tmpfs 16456152 0 16456152 0% /sys/fs/cgroup

 

[dcsapak]

maybe its the inotify watches? see https://bugzilla.proxmox.com/show_bug.cgi?id=1042

 

[Leon Roy]

it's the weirdest thing, I ran:

:~# lsof -K | grep inotify | wc -l
93

Which is hardly enough watches to saturate the system. I then shutdown all containers except the one experiencing issues. It still wouldn't come up in the Proxmox console or allow SSH in.

I noticed on the affected system SSH is in an inactive/dead state despite being set to enabled.

Trying to start it took ages, but when it did start the Console came up in Proxmox UI and I was able to SSH in.

Only differences with the affected container vs other working containers are:
1. It's using the ID of a container which I previously deleted.
2. It's Debian 9, other containers are Ubuntu 18.04.

I've just setup an Ununtu 18.04 container with identical config and same software as the affected container, everything works perfectly.

Here I was pushing myself to use Debian more vs Ubuntu but yikes, got bit. I'm curious what the issue might be since I'd prefer to use Debian long term over Ubuntu. Any ideas?