CMS R12 on SunBlade 150 _ Disc 99% full warning

[Nick44]

Hi All,
I have a customer with an R12 CMS, which has been left running and connect to an ACM2 without any use (inc backups or routine re-starts)for 4 years.
They now want to use the system again, and have found the disc 99% full message.
I have restarted the Sun box, and run a cms backup to tape. I have tried to run a cmsadm backup, but while the backup runs it causes the Sun box to crash and restart. On restart I get the error message "unable to write crash file 101 Mg needed 1 Meg available" so I guess that it not the CMS partition that's full, but the root partiton.

Have any of you good people any suggestions? We are not interested in the data that is stored, but do want to keep the agents/VDN's/trunks etc.

Thanks,
Nick.

Speak softly and carry a big switch

 

[4merAvaya]

Did you do a structured shutdown (shutdown -y -g0 -i6)? You may want to check some of the mail files: /usr/mail.

Kevin

 

[Nick44]

Thanks 4mer, already tried the structured shutdown. Whats to check on the mail files please? delete them all?


Speak softly and carry a big switch

 

[4merAvaya]

As root, go to the /usr/mail directory and see what's there. UNIX has a habit of generating messages for a lot of system functions (uucp, lp, cron, etc.). If these files are not cleared on a regular basis they can get very big.

Kevin

 

[Nick44]

Found KB01023982 on the Avaya site and followed the instructions.
We’ve managed to finish the long commands and we’ve now got it down to 32%.
Our leaps of joy were short lived as the system crashed shortly afterwards and then again when we performed a backup.
We’ve not managed to do a backup yet.

The big files were in the /var/crashes/ structure. (in a further sub-directory) and were called vmcore.1, vmcore2 etc (up to vmcore.34) there were also unix.1 and unix.2 files also. The vmcore files were 85 MB – 110 Mb and the unix.1 files were about 8 Mb I think.

Anyway we can now rule the instability of the system out from the disk space issues. (crashing on cmsadm backup)

I’m just wondering what are options are here? Whether to some how rebuild the system or troubleshoot the crashes (I guess using the new vmcore.35 file)

Speak softly and carry a big switch