Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 17 May 1996 12:16:07 -0400 (EDT)
From:      "Eric H. Jones" <ejon@zipnet.net>
To:        freebsd-stable@freebsd.org, freebsd-questions@freebsd.org
Cc:        wink@zip0.zipnet.net (Jim Winkleman), ejon@zip0.zipnet.net (Eric H. Jones)
Subject:   Series of (VM-related?) crashes on 2.1-stable
Message-ID:  <199605171616.MAA29232@zip0.zipnet.net>

next in thread | raw e-mail | index | archive | help
Greetings BSDers,

	We're running a soon-to-be-huge ISP on FreeBSD-stable
(last SUP roughly February), and have suddenly started seeing
ugly crashes on our news server.  I'm wondering if anybody else
has seen similar.

FYI, the machine is a P133 with 64 MB, scads of swap, about 28 GB
of disk on two controllers (Adaptec 2940), SMC 100baseT card,
vanilla Trident VGA card.  The disks are Quantum 34300's plus
one Compaq M1606S (well that's what it calls itself!)

My first reaction was that there may be a problem with the
VM code, but now I'm starting to think that we might be
seeing a memory error up in the pages that the kernel
keeps locked down for its own use.  Opinions?

Below is what info I've been able to glean from the messages
files:

May 2 crash:
 /kernel: kernel page directory invalid pdir=0x3d5e023, va=0xefbfe000
 /kernel: panic: invalid kernel page directory
 /kernel: 
 /kernel: syncing disks... FreeBSD 2.1-STABLE #0: Wed Mar 27 07:10:02 EST 1996

 ---------------------
May 8 crash:
 innd: newsie.dmc.com connected 26
(8 minute delay with no messages)
 /kernel: FreeBSD 2.1-STABLE #0: Wed Mar 27 07:10:02 EST 1996

 ---------------------
May 10 crash 1:
 /kernel: panic: unwire: page not in pmap
 /kernel: 
 /kernel: syncing disks... FreeBSD 2.1-STABLE #0: Wed Mar 27 07:10:02 EST 1996

 ---------------------
May 10 crash 2 (about 1 hr after crash 1):
 /kernel: 
 /kernel: 
 /kernel: Fatal trap 1

 (for some reason the reboot messages are missing here)
 ---------------------
May 17 crash:
 /kernel: 
 /kernel: 
 /kernel: Fatal trap 12: page fault while in kernel mode
 /kernel: fault virtual address	= 0xfffbeff8
 /kernel: fault code		= supervisor write, protection violation
 /kernel: instruction pointer	= 0x8:0xf01945e4
 /kernel: code segment		= base 0x0, limit 0xfffff, type 0x1b
 /kernel: 			= DPL 0, pres 1, def32 1, gran 1
 /kernel: processor eflags	= interrupt enabled, resume, IOPL = 0
 /kernel: current process		= 17105 (sh)
 /kernel: interrupt mask		= net tty bio 
 /kernel: panic: page fault
 /kernel: 
 /kernel: syncing disks... FreeBSD 2.1-STABLE #0: Wed Mar 27 07:10:02 EST 1996




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199605171616.MAA29232>