Eric Sproul wrote:
Greg Price wrote:
Try running savecore ASAP. I don't know why you didn't get a dump (and
that in itself should be investigated)
Thanks Greg,
At the time of your message, the system had already been running for 36
hours, so I don't think anything useful would have come of that. I did
find out why savecore didn't work during the reboot-- /var is not large
enough, as this machine has 16GB of RAM, and /var is only about 9GB.
Obviously a pretty big oversight in the setup. My question now is, can
I put /var/crash on its own filesystem, and will savecore find it at
boot time? Or does it just mount /var and drop it into a subdir
crash/<hostname>?
Hi Eric,
did you actually run the savecore command, or are you just making an
assumption about what it would have resulted in? Making assumptions
can be detrimental.
If you rtfm for savecore and dumpadm you'll see that
* you can run savecore and point it at pretty much any directory you
have write permissions for
* dumpadm lets you specify both the savecore directory and a dedicated
dump device.
James C. McPherson
--
Solaris kernel software engineer, system admin and troubleshooter
http://www.jmcp.homeunix.com/blog
Find me on LinkedIn @ http://www.linkedin.com/in/jamescmcpherson
_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss
