After upgrading to 12.2+, Agents are now taking 4-8 minutes to start when attempting to execute a job.
Design change in the job messenger can cause modified HEADER.* files to send improper messenger commands, causing a timeout to occur before the Agent attempts to connect via different methods.
Release : 12.2
Component : AUTOMATION ENGINE
If this occurs, please check the HEADER.* for the appropriate OS; for example, HEADER.UNIX. Check to see if one is present in the client affected. If so, compare this to the header in client 0.
If your pre-upgrade header is different than the default, this can cause the issue.
In the version of this seen, the HEADER file had the IP address not passed:
: set &UC_IP_ADDR = ''
This worked in versions lower than 12.2, but within 12.2+ caused the Agent timeout.