The box became unresponsive, so I had to reboot it.
Quote:
Perhaps you should also get the header line.
|
Sorry, here's the header too (after reboot)
Code:
[root@localhost ~]$ ps aux | head -n1
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
[root@localhost ~]$ ps aux | grep -i networkmanager
root 477 0.1 3.4 447909 8237 ? Ssl 12:51 0:00 /usr/sbin/NetworkManager --no-daemon
So, after reboot, VSZ = vsize (in KiB) = (447909 KiB) / 1024 = 437 MiB.
And free
Code:
[root@localhost ~]# free -m
total used free shared buff/cache available
Mem: 237 62 15 4 159 140
Swap: 1023 0 1023
^ That looks good. Of course, following the reboot, the box is running much smoother now, but--before the reboot--it was heavily swapping. I'm not worried about buffers/cache, but when the box was swapping and performance was horrible, NetworkManager was the process holding the most RAM on this machine.
Is this a known issue with NetworkManager on Cent7? This is a headless, CLI-only server with only 256M of RAM and no GUI. NetworkManager is now the default networking daemon for server installs of CentOS 7, and I'm just trying to understand why it's eating all my RAM. I didn't have this issue with Cent 6.