Re: Unidentified subject!

Eric Smith (eric@brouhaha.com)
8 Nov 1996 20:25:59 -0000

"Scott D. Webster" <scottw@frontier.wilpaterson.edu> wrote:
> I'm having a little trouble with the NCR53c8xxx driver on two
> EB164's (running RHL 3.0.3).
[...]
> The problem only seems to manifest itself when the system crashes
> and fscks on reboot. It will sometimes completely hang the system. The
> strange thing is that if I reboot to the origional 1.3.90 RedHat kernel,
> it will not hang. Any idea what I've set wrong that would cause this?

I've had the same problems on a UDB running RH 3.0.3 every time I've tried to
upgrade to any 2.0.x kernel, using either the 8xx driver or the old 53c7,8xx
driver.

I upgraded one of my UDBs to RH 4.0, thinking that the problem must surely
be fixed, but no, it still happens.

However, I've seen a lot of other SCSI related faults as well.

Before everyone suggests again that I have bad cabling, termination, or
broken hardware, I'll point out that
1) I've checked all those things very carefully, even to the extent of
verifying SCSI signal integrity on an oscilliscope
2) The 1.3.90 kernels (both Red Hat's and my own builds) are rock solid
on the same machine.

For the time being, I'm sticking with 1.3.90 on my production system.
Unfortunately I've already upgreaded my other UDB to RH 4.0 with kernel
2.0.18. Since it's ELF-based, there's no way I can go back to 1.3.90 without
reinstalling RH 3.0.3.

I've not yet had time to do a binary search and figure out what kernel version
broke things.

Eric

--
To unsubscribe: send e-mail to axp-list-request@redhat.com with
'unsubscribe' as the subject.  Do not send it to axp-list@redhat.com



Feedback | Store | News | Support | Product Errata | About Us | Linux Info | Search | JumpWords
No Frames | Show Frames

Copyright © 1995-1997 Red Hat Software. Legal notices