What happened?
Friday night the media servers started reporting error 96 [Status = unable to allocate new media for backup, storage unit has none available] and three drives were downed.
Saturday morning I determined that all of the media servers did not have access to any drives, only the master server was performing any backups.
In the admin console, under “media and device management”, under “hosts”, it was showing the master server and one other server that I have been using for testing the upgrade from 4.5 to 5.1 earlier in the week. none of the other media servers were listed.
In the end, I had to delete all of the associated drives on the media servers and add them back in through the gui on the master in order for the media severs to sync with the master.
the test server was built from a imported catalog tape generated a week ago. The test server had a different name and ip but on the same subnet. The original bpconf and vm.conf were intact on the test server and nis had been left running.
Any ideas on how / why the master server would have picked up the test server and at the same time drop the media servers? Has anyone ever experienced anything like this?
Running netbackup 4.5 fp5
Master server running solaris 8.
Backup sans, two brocade switches, l700 library.
Ibm lto gen 1 fc drives.
All media servers running sso and at solaris 9.
one media server running windows 2003 and sso as well.
Thanks.
Friday night the media servers started reporting error 96 [Status = unable to allocate new media for backup, storage unit has none available] and three drives were downed.
Saturday morning I determined that all of the media servers did not have access to any drives, only the master server was performing any backups.
In the admin console, under “media and device management”, under “hosts”, it was showing the master server and one other server that I have been using for testing the upgrade from 4.5 to 5.1 earlier in the week. none of the other media servers were listed.
In the end, I had to delete all of the associated drives on the media servers and add them back in through the gui on the master in order for the media severs to sync with the master.
the test server was built from a imported catalog tape generated a week ago. The test server had a different name and ip but on the same subnet. The original bpconf and vm.conf were intact on the test server and nis had been left running.
Any ideas on how / why the master server would have picked up the test server and at the same time drop the media servers? Has anyone ever experienced anything like this?
Running netbackup 4.5 fp5
Master server running solaris 8.
Backup sans, two brocade switches, l700 library.
Ibm lto gen 1 fc drives.
All media servers running sso and at solaris 9.
one media server running windows 2003 and sso as well.
Thanks.
