restart keepalived donot work, Can't open PID file /var/run/keepalived.pid (yet?) after start: No such file or directory about keepalived HOT 1 CLOSED

I have tested systemctl restart keepalived on my CentOS 7 VM and it works fine for me.

There are some strange things about the log output

1. systemd logs "Stopped ..." before it logs "Stopping ..."
2. The new keepalived logs that keepalived is already running.

Point 2 means that keepalived has found an old PID file and the process whose PID is in the file still exists. Each of the keepalived processes (parent process, VRRP process checker process) creates a PID file and each of them is checked. It is presumably the VRRP or checker process that is still running, since systemd complains that /var/run/keepalived.pid does not exist. After keepalived determines another instance of keepalived is still running the new keepalived then exits, which is why the keepalived.pid file does not exist.

Does this problem always happen when you restart keepalived, or is it intermittent? Have you checked the RedHat bugzilla for any relevant keepalived issues on RHEL7?

The interaction between keepalived and systemd has significantly improved since keepalived v1.3.5, as indeed has keepalived itself (there have been over 3000 non merge commits since v1.3.5). In order for us to look at the problem further, you would need to build and install the latest keepalived (i.e. v2.2.8 or later) and observe the problem occurring with that. You would also need to provide the full logs for keepalived since it started, and the relevant systemd logs.

I am closing this issue now, but if you provide the details identified in the previous paragraph and the problem is still occurring with keepalived v2.2.8+ we will reopen the issue.

from keepalived.