Description of problem:
When using misc_check with misc_timeout, after a failed check, restarting keepalived sends a SIGTERM to systemd (pid 1)

Version-Release number of selected component (if applicable):
keepalived-2.0.10-10.el8.x86_64
8.2

How reproducible:
100%

Steps to Reproduce:
1. Launch keepalived with a config that uses misc_check and misc_timeout:

virtual_server 172.16.0.32 1234 {
    lb_algo rr
    lb_kind NAT
    protocol UDP
    delay_loop 5
    delay_before_retry 5
    retry 3

    real_server 172.16.0.32 1235 {
        weight 1
        uthreshold 100000
        MISC_CHECK {
            misc_path "sleep 3"
            misc_timeout 2
        }
    }
}

Logs:

Tue May 19 08:15:03 2020: Starting Keepalived v2.0.10 (11/12,2018)
Tue May 19 08:15:03 2020: Running on Linux 4.18.0-193.el8.x86_64 #1 SMP Fri Mar 27 14:35:58 UTC 2020 (built for Linux 4.18.0)
Tue May 19 08:15:03 2020: Command line: 'keepalived' '-D' '-l' '-G' '-n' '--pid=keepalived.pid' '-f' 'keepalived.conf'
Tue May 19 08:15:03 2020: Opening file 'keepalived.conf'.
Tue May 19 08:15:03 2020: Starting Healthcheck child process, pid=857707
Tue May 19 08:15:03 2020: Opening file 'keepalived.conf'.
Tue May 19 08:15:03 2020: WARNING - default user 'keepalived_script' for script execution does not exist - please create.
Tue May 19 08:15:03 2020: WARNING - script `sleep` resolved by path search to `/bin/sleep`. Please specify full path.
Tue May 19 08:15:03 2020: SECURITY VIOLATION - check scripts are being executed but script_security not enabled.
Tue May 19 08:15:03 2020: Initializing ipvs
Tue May 19 08:15:03 2020: Gained quorum 1+0=1 <= 1 for VS [172.16.0.32]:udp:1234
Tue May 19 08:15:03 2020: Activating healthchecker for service [172.16.0.32]:udp:1235 for VS [172.16.0.32]:udp:1234


2. Wait for misc_check to fail

Tue May 19 08:15:10 2020: Misc check for [[172.16.0.32]:udp:1235 VS [172.16.0.32]:udp:1234] by [/usr/bin/sleep] timed out.
Tue May 19 08:15:10 2020: Removing service [172.16.0.32]:udp:1235 to VS [172.16.0.32]:udp:1234
Tue May 19 08:15:10 2020: Lost quorum 1-0=1 > 0 for VS [172.16.0.32]:udp:1234


3. Send a TERM signal to keepalived (or systemctl restart)

$ sudo kill -TERM $(cat keepalived.pid)


Actual results:
systemd receives a TERM signal from keepalived and restarts

May 19 08:15:13 undercloud-0.redhat.local systemd[1]: Received SIGTERM from PID 857707 (keepalived).
May 19 08:15:13 undercloud-0.redhat.local systemd[1]: Reexecuting.
May 19 08:15:13 undercloud-0.redhat.local kernel: printk: systemd: 49 output lines suppressed due to ratelimiting
May 19 08:15:13 undercloud-0.redhat.local systemd[1]: systemd 239 running in system mode. (+PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ >
May 19 08:15:13 undercloud-0.redhat.local systemd[1]: Detected virtualization kvm.
May 19 08:15:13 undercloud-0.redhat.local systemd[1]: Detected architecture x86-64.

Note that "systemctl restart" returns an error


Expected results:
keepalived should exit gracefully, systemd shouldn't be affected


Additional info:

https://github.com/acassen/keepalived/pull/1478/commits/2738ee60b394ab859de18d1a4c8e332cc55887ea fixes that issue.