Bug 1846665

Description of problem:
I first reported this in bug 1803794 and was asked to create a new bug.  The problem is that the /etc/cron.daily/rhsmd script it not killing off the rhsmd process that it is supposed to kill off.  Here are the lines of the script that are the issue:

 12 /usr/libexec/rhsmd -s &
 13 sleep $rhsmd_timeout;
 14
 15 ps_check=$(pgrep -f '/usr/libexec/rhsmd')
 16 if [ -n "$ps_check" ]; then
 17   pkill -f '/usr/libexec/rhsmd' >/dev/null 2>&1
 18   logger -t rhsmd -p user.warn "rhsmd process exceeded runtime and was killed." >/dev/null 2>&1
 19 fi

On line 12 the rhsmd process is started and then the script sleeps for the process timeout amount from the config file.  What we have observed is that the rhsmd process exits in just a few seconds typically.  After the sleep the script looks for a /usr/libexec/rhsmd process and if it finds one it kills it.  It doesn't do any validation that the rhsmd process it is killing is the one it started initially which is the intent.  Instead it kills the process and then in all of our 7.8 systems we now see the following in the syslog output:

Jun  5 03:26:02 testhost1 rhsmd: rhsmd process exceeded runtime and was killed.
Jun  5 03:26:05 testhost1 dbus[778]: [system] Activating service name='com.redhat.SubscriptionManager' (using servicehelper)
Jun  5 03:26:05 testhost1 dbus[778]: [system] Successfully activated service 'com.redhat.SubscriptionManager'



Version-Release number of selected component (if applicable):
subscription-manager-1.24.26-3.el7_8.x86_64

How reproducible:  This happens on our RHEL 7.8 systems everyone morning in the logs.


Steps to Reproduce:
1.Wait until script runs
2.Observer output in the log files.
3.

Actual results:


Expected results:
I would expect the script would track the pid of the rhsmd process it starts and only kill that process off rather than blindly killing off all rhsmd processes on the system.


Additional info: