Showing results for 
Search instead for 
Did you mean: 

Netbackup Appliance 5240 upgrade 3.0 to 3.1

Level 2

Goodmorning 2 ya all

I updated to version 3.1 and works fine.. but we encounter some problem regarding the update. I would like to tell about my experience about this here. Short version: 

Error: "Wont reboot"

Solution: manual reboot

OK so I've uploaded the 3.1 one binary (md5 checked it before). This through the shell (ssh) with the sysadmin account. We have two boxes.. I did this on both at same time (should take one at the time). I bpstop the master server but should do that then the script didnt find my master and stopped. So manually stopped the SLP exclude the date for the actual time in the schedule, wont have any interference on our 5240:s. (here Vertias should develeoped some script to stop enough items on master so that upgrade of deduplication nodes could be done.)

Started the update install <file> enter and prechecked went good, the guide tells that there will be severeal reboots but (actually three). OK so after waiting for 2.5 hours.. we waited for 1 hour more.. Now something was wrong. 

  • No ping to the boxes. 
  • We could access them through the mgm, Mangement(so check this before the update)
  • We had recived the mail from the appliance that the update  went successfull
  • We contact the Veritas support and concluded that reboot manually. 

OK now we had checked that the box was in some idle status. And this what we did to get it to work:

  • Go to mgm (mangement) and chose reset on the power so they rebooted
  • Let it boot and let it hang, wait 15minutes then
  • Go to mgm and chose reset on the power reboot
  • Done, verify it on NBU master and see that you have the boxes there
  • ? Other that have same problem, did it requier 2 reboots
  • ? Only 5240 have seen simliar to the 5230.
  • ? Some technical info regarding this. 

Regards Pontus Ankarbåge


Level 4

We are seeing the same behaviour with one of our 5240 appliances as well, since the 3.1 update.  And like your are describing, it takes two resets.  If you look on the console, you will probably see the kernel panic.  We have had a case open for a while.

There is a tech note for kernel panics with certain conditions and 10GigE cards, but no solution yet:

There is another tech note for kernel panics attributed to FC/SAN connections, this one has a hotfix provided:

We also experienced the 'spoold' daemon crash problem after the 3.1 update, which was a major pain, at least there is a fix for that one: