Source: lxcfs Version: 2.0.7-1 Severity: normal Tags: upstream As reported here: https://github.com/lxc/lxd/issues/3517 which points to https://github.com/lxc/lxcfs/issues/164 (where is stated that virtualizing the btime field has been reverted due to https://github.com/lxc/lxcfs/issues/189), I'm experiencing wrong stimes for processes.
As I find having accurate start times of processes far more important than having an accurate uptime counter for my containers, as do other people as well apparently, I would like to see this reverted until a better solution for the btime field /uptime is found. # lxc-attach -n bla bla# ls -l /run/apache2/apache2.pid -rw-r--r-- 1 root root 5 Dec 27 14:15 /run/apache2/apache2.pid bla# ps -fp `cat /run/apache2/apache2.pid` UID PID PPID C STIME TTY TIME CMD root 5691 1 0 17:29 ? 00:00:01 /usr/sbin/apache2 -k start bla# uptime 21:35:21 up 3 days, 2:57, 0 users, load average: 0.15, 0.13, 0.17 bla# exit host# uptime 21:35:33 up 3 days, 7:00, 5 users, load average: 0.13, 0.13, 0.16 You can see the difference in time between starting the host and the container is the difference between the time shown by ps and the actual start time of the process. Thanks, Paul -- System Information: (reported from a different system as the LXC system has no email as yet)