Is this a newly installed cluster that has never worked before?
Or has this worked before and is now failing?
Have you checked disk space on NBU catalog volume?
NBU will shutdown emm if catalog disk gets to 99% full.
If this is a production server that has worked fine previously, do not attempt an upgrade at this stage.
You need to always perform a catalog backup before an upgrade and will not be able to do so with NBU not running.
To troubleshoot, please perform the following steps:
- Use bpps -x to check for remaining NBU processes. If any, stop them with 'netbackup stop'.
Leave pbx running.
- Clear the faulted resouces on both nodes.
- Online all resources on one node except for the NetBackup resource (nbu_server)
(In other words: diskgroup, volume(s), mount point(s) and IP)
- Check that virtual IP resolves to master virtual hostname:
bpclntcmd -ip <virtual-IP>
- Create log folders (if they don't exist already) on this node under /usr/openv/netbackup/logs:
bpdbm bprd admin
- Freeze the SG
- Manually start NBU from command line:
netbackup start
- Carefully check for errors
- Check processes from another window with 'bpps -x'
Please post output of 'netbackup start' as well as 'bpps -x' plus these 2 commands:
nbemmcmd -listhosts -verbose
nbemmcmd -getemmserver
If emm does not remain running, post output of emm log:
vxlogview -o 111 -t 00:10:00
(above command will report on last 10 minutes. Ensure -t value includes enough time to catch the last startup attempt.)
Depending on errors and/or output of bpps, we may need to have a look at the other logs as well.
