I can now "reliably" get my Shrike machine to kernel panic when running the stock /etc/crond.daily/slocate.cron script. This was discovered after the machine began crashing at 04:02:02 *every* night. The script: #!/bin/sh renice +19 -p $$ >/dev/null 2>&1 /usr/bin/updatedb -f "nfs,smbfs,ncpfs,proc,devpts" -e "/tmp,/var/tmp,/usr/tmp,/afs,/net" This is but the latest in a fairly unbelievable series of problems I have had with RH9. This particular problem actually began in the early days of the installation, but seems to have worsened over time, that is, initially it would crash about once per week running updatedb, now it crashes every night (as of now I no longer run crond, so the problem has been temporarily stayed). As a result of one of the crashes the machine required manually running fsck on both major partitions (/ and /home) after it was power-cycled. I'm running ext3. This machine is about two years old, and is built with fairly high end components: AMD Athlon 1100 Asus KT133 Motherboard 768M Crucial ECC memory Matrox Millennium II Adaptec 29160N SCSI adapter Seagate Cheetah X15 18GB SCSI-3 Hard Drive Pioneer 12X SCSI DVD Yamaha 8/4/24 SCSI CDR Sound Blaster X-Gamer? 5.1 Intel 10/100 NIC This computer has pretty much never crashed in the two years prior to Red Hat 9 being released. It ran RH8 without a reboot for six months, being used everyday as a developer workstation. I generally regard the system as being very stable. All instability issues started when RH9 was installed. I have experienced similar stability issues another RH9 machine that also worked perfectly with RH8. If anyone has any ideas as to what might be the problem here I'd greatly appreciate hearing about them. Will RedHat's 90-day installation support cover these issues if I were to purchase a retail box of the software and install it? Best regards --Tod Liebeck ------------------------------------------------- This mail sent through IMP: http://horde.org/imp/
