The Unix Agent stops upon restart when jobs are still running in the background

Products

• CA Automic Workload Automation - Automation Engine

Issue/Introduction

When a Unix Agent is stopped and started while there were still jobs running, it doesn't always come back up properly.

The exact behavior can be different depending on how the Agent is started:

• When the Agent runs under control of the Service Manager, it restart and stops in a loop
• When started manually or using a systemd script, it may stop again or startup correctly

When you run a tcp/ip=9 trace on the Agent (set in the Agent's ini-file), you will find the following exception:

20260212/150726.266 - 25               UserServiceDriverProcessMonitor [pid=28155], no message, reason: java.io.EOFException
20260212/150726.266 - 25               checkProcessExitStatus()
20260212/150726.270 - 25               checkProcessExitStatus: Child process update received: SigInfo [si_signo=17, si_errno=0, si_code=2] SigChildInfo [si_pid=28155, si_uid=0, si_sigval=6]
20260212/150726.271 - 38     U00009909 TRACE:Rcv 0 bytes from UserServiceChannel{fd=7} size 000008
20260212/150726.271 - 38               00000000  00000000 00000000                    >........<
20260212/150726.272 - 38               Unable to info for process (28004)
20260212/150726.272 - 38     U00045014 Exception 'java.io.IOException: "Unable to info for process (28004)"' at 'com.automic.os.process.unix.userservice.UserServiceDriver.receiveMessage():541'.
20260212/150726.272 - 38     U00045015 The previous error was caused by 'java.io.EOFException: "null"' at 'com.automic.os.process.unix.userservice.UserServiceChannel.readAll():102'.

When setting job_debug=9 the issue doesn't happen.

Environment

Agent Unix

Cause

DE182056

Resolution

This will be fixed in a future version of the product.