This is not a question but something that I have come across that I wanted to share.
While performing client upgrades from my Solaris 8.1.2 master I would see the following in the Detailed Status window:
Sep 5, 2019 9:56:25 AM - Info RUNCMD (pid=7562) Service shutdown successful.
Sep 5, 2019 10:56:46 AM - Info RUNCMD (pid=7562) Execution of the nbinstallagent on the remote host failed
Sep 5, 2019 10:56:46 AM - Info RUNCMD (pid=7562) exiting with status: 7207
As you can see this would run for 1 hour (3600 seconds) and then abort / timeout. It turns out that there is an undocumented variable that can be added to the bp.conf file on the master server that can be modified to adjust the time out. The variable is called VXUPDATE_CLIENT_READ_TIMEOUT_SECONDS and the default value is 3600. I needed to modify the value to be 10 hours (36000 seconds) so I added the line
VXUPDATE_CLIENT_READ_TIMEOUT_SECONDS = 36000
to bp.conf (no need to restart any processes) and now my upgrades complete successfully. This was particular important on AIX servers as they have to traverse the whole process tree and search each process for specific NetBackup paths to verify that nothing is using them before performing the upgrade.
Note: you will need to restart the NetBackup processes on the client server after getting the 7207 error as the timeout prevents the script from restarting them.
I hope this helps anyone who might be having the same issue.
Regards.
