proc.num issues with 5.4.2 and Zabbix Agent 2 (active check)

[tjh]

Since upgrading to 5.4.2 this morning, I'm having trouble with a proc.num on two hosts.

Every now and then (there seems to be no schedule to it) my zabbix_agent2 will return 0 for the number of running processes. Then a minute or two later, it'll return to normal.
I'm getting for a ntpd check on one host (just proc.num[ntpd]) and an Apache2 check on another host.

Prior to the 5.4.2 upgrade this morning, these have never ever alarmed before. I've verified that the processes aren't restarting etc. Nothing else is alarming/having issues, just one proc.num check on each host.

Wondering if anyone else is seeing similar?

Thanks!

[tjh]

So I just found this which is a different implemention of proc.num in Zabbix Agent 2. I suspect this is the cause of my problems, I have added a note to the original ticket.

[tjh]

It seems another person is experiencing this too, they have logged a bug here.