After an upgrade of the Linux master server to NetBackup 8.1 the NetBackup Java administration console encounters performance issues. The GUI may take a couple minutes or more to come up and attempts to view the activity monitor backup policies, etc... are slow to display.
Viewing the system log (/var/log/messages) showed multiple entries similar to the following:
Feb 19 10:06:36 dcmaster1 xinetd: START: bpcd pid=13267 from=::ffff:10.101.18.20
Feb 19 10:06:36 dcmaster1 xinetd: EXIT: bpcd status=0 pid=13267 duration=0(sec)
Feb 19 10:28:04 dcmaster1 xinetd: START: vnetd pid=19856 from=::ffff:10.101.18.20
Feb 19 10:28:04 dcmaster1 xinetd: EXIT: vnetd status=23 pid=19856 duration=0(sec)
Notice that bpcd and vnetd appear to be started by xinetd.
The vnetd log also showed bind failures indicating it was having connectivity issues:
10:08:37.013 [16171.16171] <2> legacy_listen: bind(13724) failed: 98
10:08:38.014 [16171.16171] <2> legacy_listen: bind(13724) failed: 98
The NetBackup bpcd and vnetd processes were still configured in xinetd. Checking the directory on the master server:
showed there were configuration files for bpcd and vnetd. Prior to version 7 NetBackup did make use of the operating system xinetd process to manage the starting of these two processes. Now NetBackup starts bpcd and vnetd as daemons. Having the processes attempt to be started by xinetd was interfering with the already running daemons and caused delays trying to update the administration console.
The NetBackup configuration files for bpcd and vnetd can be safely removed from the /etc/xinetd.d directory. Restart xinetd to commit the change by running
service xinetd restart
Once that is done the administration console can be restarted and should perform much better.
Related Knowledge Base Articles
Was this content helpful?
Rating submitted. Please provide additional feedback (optional):