Dear fellow-managers!
Thanks to
Jim Belonis (belonis_at_dirac.phys.washington.edu)
Alan Rollow (alan_at_nabeth.cxo.dec.com)
David Summers (dave_at_tetricus.freeserve.co.uk)
It took me quite a bit of time to solve the problem, the main reason
being the users running batch jobs around the clock. But on Saturdays
most of the users won't create batch jobs, so, over the weekends there
was enough time left for me ... :-(.
First of all I followed the advice given by sys_check and adjusted the
virtual memory parameters. This helped for one disk on either machine.
Further investigation (by defragment -V <domain>) revealed that very
large files (on the order of 700 MB in my case), which in turn have lots
of extents, can kill defragment. The VERY simple (I must admit) solution
was to move these files to some other disk and then to copy them back.
Everything has been fine ever since. As a consequence, I run defragment
jobs regularly now.
-----------------------------------------------------------------------
Original message follows:
Hello managers!
I'm having a problem defragmenting on some advfs disks. The utility
would start working but after a while I get console messages telling me
something about malloc failures. After a dozen (or so) messages of that
kind the system crashes with "AdvFS exception ..." and writes a crash
dump.
This behaviour - which is reproducible - occurs on four of my disks. The
other disks (of the same type!) are ok. I'm running an Alphastation 200
4/166 with Unix 4.0B, patch level 8 and a PW433au with Unix 4.0C, patch
level 5. The disks are connected to SC200 controllers (NCR 810 chip) on
either machine.
Is there any help for me out there?
--
Dipl.-Phys. Rolf-Peter Kienzle
Dept. of Remote Sensing and Land Information Systems
University of Freiburg, D-79085 Freiburg, Germany
Tel/Fax +49 761 203 8643/8640 e-mail: kienzlep_at_uni-freiburg.de
Received on Thu Apr 22 1999 - 15:41:01 NZST