Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 14 Sep 2000 11:28:34 +0930
From:      Greg Lehey <grog@lemis.com>
To:        kts <bangel@elite.net>
Cc:        hackers@FreeBSD.ORG
Subject:   Re: help debugging kernel crash dump
Message-ID:  <20000914112834.C45769@wantadilla.lemis.com>
In-Reply-To: <Pine.BSD.3.91.1000913040827.6020A-100000@almond.elite.net>; from bangel@elite.net on Wed, Sep 13, 2000 at 04:12:14AM -0700
References:  <Pine.BSD.3.91.1000913040827.6020A-100000@almond.elite.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday, 13 September 2000 at  4:12:14 -0700, kts wrote:
> Hi
>
> Just wanted to see if anyone could give me some feedback as to what
> happened, or how to get more info. My machine was just sitting on
> the LAN and it panic'd. X was running and it had some ssh's open.
>
> I'm running 4-STABLE from August 3rd.
>
> IdlePTD 3522560
> initial pcb at 2ce300
> panicstr: page fault
> panic messages:
> ---
>
> dumping to dev #wd/0x20001, offset 278560
> dump ata0: resetting devices .. done
> 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 48 47 46 45 44 43 42 41 40
> 39 38 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16
> 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
> ---
> #0  0xc0141ef0 in boot ()
> (kgdb) bt
> #0  0xc0141ef0 in boot ()
> #1  0xc014228d in panic ()
> #2  0xc025b166 in trap_fatal ()
> #3  0xc025ae19 in trap_pfault ()
> #4  0xc025a9df in trap ()
> #5  0xc01acb69 in fr_makefrip ()
> #6  0xc01b2a98 in fr_checkicmpmatchingstate ()
> #7  0xc01b2cd9 in fr_checkstate ()
> #8  0xc01ad758 in fr_check ()
> #9  0xc01a17e8 in ip_output ()
> #10 0xc019fa6a in icmp_send ()
> #11 0xc019f9eb in icmp_reflect ()
> #12 0xc019f314 in icmp_error ()
> #13 0xc01aa46e in udp_input ()
> #14 0xc01a0498 in ip_input ()
> #15 0xc01a050b in ipintr ()
> #16 0xc024de85 in swi_net_next ()
> Cannot access memory at address 0x531b6b24.
> (kgdb)
>
> any way I can get more information from that? I know using a kernel
> compiled with -g would help, but how would that be done, because
> I've been sup'ing my src every day...

Yes, a kernel with -g would be a great help.  The stack trace suggests
a network problem, and possibly somebody with more experience in this
area would recognize this signature.

To build a kernel with -g, you either run config with the -g option,
or better put this into the config file:

  makeoptions	DEBUG=-g		#Build kernel with gdb(1) debug symbols

'make install' will still install a stripped version, but you'll have
the unstripped version to use for crash dumps.

Unfortunately, you can't just build a debug kernel from these sources
and expect to get the same thing; it seems that there are some minor
differences in code generation when you set -g.  It's possible,
though, that there would be enough similarity that it would help.

Greg
--
Finger grog@lemis.com for PGP public key
See complete headers for address and phone numbers


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000914112834.C45769>