So I built a cluster that I was trying to use only LXC containers for the sites that I am hosting. I am guessing I will have between 40-50 sites per host, however I am getting lots of strange issues. the Containers are acting strange (giving lots of "to many files" open errors). The strange thing though on the HOST system I modified the values to be:
[ /etc/security/limits.conf ]
* soft nofile 565000
* hard nofile 565000
root soft nofile 565000
root hard nofile 565000
pve-manager/5.2-9/4b30e8f9 (running kernel: 4.15.18-5-pve)
Example of logs in /var/log/messages on host:
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
syslog:Sep 24 20:26:01 pve01 pvesr[2411244]: Unable to create new inotify object: Too many open files at /usr/share/perl5/PVE/INotify.pm line 396.
Even after rebooting the "host" and starting the containers I continue to get "to many files" open errors. What is the suggest optimization to change on a Proxmox host in order to get containers to work without issues?
Running "lsof | wc -l" on the main host returns:
224973
[ /etc/security/limits.conf ]
* soft nofile 565000
* hard nofile 565000
root soft nofile 565000
root hard nofile 565000
pve-manager/5.2-9/4b30e8f9 (running kernel: 4.15.18-5-pve)
Example of logs in /var/log/messages on host:
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
daemon.log.1:Sep 22 22:16:33 pve01 systemd[1]: inotify_init1() failed: Too many open files
syslog:Sep 24 20:26:01 pve01 pvesr[2411244]: Unable to create new inotify object: Too many open files at /usr/share/perl5/PVE/INotify.pm line 396.
Even after rebooting the "host" and starting the containers I continue to get "to many files" open errors. What is the suggest optimization to change on a Proxmox host in order to get containers to work without issues?
Running "lsof | wc -l" on the main host returns:
224973
Last edited:
