I was having a problem with an Ultra/2 machine randomly crashing for no
apparent reason. The messages file would show BAD TRAP errors on one of the
dual 300mhz cpus. The problem was that there was a bad batch of cpu's for
the Ultra/2s. And we had one of them. When i posted the question here I was
immediatly mailed by about 20 people who had had the same problem, so I was
pretty sure it was the bad cpus. Sun sent me two new cpus, and I poped them
in and haven't had any more problems. Thanks for all of your replys, it
saved me tons of time in figuring the problem out, well without it, I would
have never guess bad cpus. Sun tech. immediatly aggreed that the problem
was bad cpus.
The stranged thing I thought was that the machine ran fine for about 5
months before completly going bonkers. Once it crashed for the first time,
it wouldn't run stable for anything. Thanks again for everyone who replied.
> I have been having some real strange random crashes on an Ultra/2 machine.
>Its running Solaris 2.6. I been trying to narrow down what the problem is,
>of course, and there is one message in the /var/adm/messages file that I do
>not recognize. Anyone see this before? Here it is.
>Aug 3 08:28:10 samson unix: BAD TRAP: cpu=1 type=0x34 rp=0x303798b0
>Aug 3 08:28:10 samson unix: basename: alignment error:
>Aug 3 08:28:10 samson unix: addr=0xd6
>Aug 3 08:28:10 samson unix: pid=10624, pc=0x100706f0, sp=0x30379940,
>Aug 3 08:28:10 samson unix: g1-g7: 10425c00, da2b80d, 100550e8, 730, 0, 0,
>Aug 3 08:28:10 samson unix: Begin traceback... sp = 30379940
>Aug 3 08:28:10 samson unix: Called from 100704e0, fp=303799b0, args=1
>10042400 1 600295e0 3 0
>Aug 3 08:28:10 samson unix: Called from 10055150, fp=30379a18,
>args=60ac42a0 30379a78 100704b8 0 0 0
>Aug 3 08:28:10 samson unix: Called from 1002dfa4, fp=30379a80,
>args=60ac4238 30379ae0 10426504 80000001 3 0
>Aug 3 08:28:10 samson unix: Called from ef7c3f78, fp=efffeb08, args=0 2000
>5 80000001 3 0
>Aug 3 08:28:10 samson unix: End traceback...
>Aug 3 08:28:10 samson unix: panic[cpu1]/thread=0x60e862c0: trap
>This happens at random times AFTER the reboot. Any ideas would be very
>welcome. I don't know if this is the problem, because it happens after the
>machine crashes, but I don't know what it means either.
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:12:45 CDT