Yesterday I was investigating a problem reported from our monitoring: a sparc machine with Solaris didn’t responded after a power outage (BTW I wasn’t in Prague during heavy storm on Sunday).

When I connected to http interface of SP and run java serial console, I saw that the problem is with full / filesystem. Unfortunately it hangs before you get any prompt.

Now it is according Solaris 10 certification (or this bolog: 1) get OK prompt 2) boot to single user and 3) make some free space.

On ILOM sp you can connect by ssh and then:

-> set /HOST/bootmode script="setenv auto-boot? false"
Set 'script' to 'setenv auto-boot? false'

-> set /HOST send_break_action=break
c)ontinue, s)ync, r)eboot, h)alt? r
Resetting...

and then on console screen you can see that machine turns into OK prompt.

boot -s

My hint what to check first in this situation: /var/crash You can usually delete it right away without any harm and in this case it was about 8G.

Tags

Napsat komentář