General Protection Fault,

Hi,

Sorry for posting this to the wrong mailing list, but I can't seem to find
the right one on your web-page (easily that is :-)

With Squid 2.3 DEVEL 2, I get general protection faults in Linux 2.0.35 with
3 x 9Gb HDD and 256Mb RAM. Only this time, as opposed to 2.2.STABLE4, it
doesn't bring down the Linux box anymore, squid will just restart
(RunCache).

Maybe it's a bug, or what else could be wrong ? However, it doesn't seem to
happen on my 2.2.15 box... (Same config, different kernel)

Here's the output of 2 crashes in my syslog:

Aug 24 10:59:14 loxy kernel: general protection: 0000
Aug 24 10:59:14 loxy kernel: CPU: 0
Aug 24 10:59:14 loxy kernel: EIP: 0010:[tcp_recvmsg+368/1036]
Aug 24 10:59:14 loxy kernel: EFLAGS: 00010202
Aug 24 10:59:14 loxy kernel: eax: f000e987 ebx: 00000000 ecx: 0d97fc30
edx
: f000f84d
Aug 24 10:59:14 loxy kernel: esi: 0d97fccc edi: 000001a0 ebp: 0d97fc0c
esp
: 02c4fee8
Aug 24 10:59:14 loxy kernel: ds: 0018 es: 0018 fs: 002b gs: 002b ss:
001
8
Aug 24 10:59:14 loxy kernel: Process squid (pid: 5136, process nr: 8,
stackpage=
02c4f000)
Aug 24 10:59:14 loxy kernel: Stack: 0d97fc0c 02c4ff7c 00000000 00000800
046e7992
 0d97fc30 00000000 000001a0
Aug 24 10:59:14 loxy kernel: 065f6018 0d0c8f48 f46f703f 001507f6
0d97fc0c
 02c4ff78 0000fe5f 00000800
Aug 24 10:59:14 loxy kernel: 00000000 02c4ff7c 0000ffff 0d0c8f00
080e024c
 0d0c8f90 001368db 0d0c8f90
Aug 24 10:59:14 loxy kernel: Call Trace: [inet_recvmsg+114/136]
[sock_read+171/1
92] [sys_read+192/232] [system_call+85/124]
Aug 24 10:59:14 loxy kernel: Code: 8a 40 0d a8 02 74 04 ff 4c 24 10 8b 7c 24
10
39 7b 30 0f 87

~~~~~~

Aug 24 12:27:13 loxy kernel: general protection: 0000
Aug 24 12:27:13 loxy kernel: CPU: 0
Aug 24 12:27:13 loxy kernel: EIP: 0010:[tcp_recvmsg+368/1036]
Aug 24 12:27:13 loxy kernel: EFLAGS: 00010217
Aug 24 12:27:13 loxy kernel: eax: f000e987 ebx: 00000000 ecx: 0869a438
edx
: f000f84d
Aug 24 12:27:13 loxy kernel: esi: 0869a4d4 edi: 000001a1 ebp: 0869a414
esp
: 02c4fee8
Aug 24 12:27:13 loxy kernel: ds: 0018 es: 0018 fs: 002b gs: 002b ss:
001
8
Aug 24 12:27:13 loxy kernel: Process squid (pid: 5929, process nr: 8,
stackpage=
02c4f000)
Aug 24 12:27:13 loxy kernel: Stack: 0869a414 02c4ff7c 00000000 00000800
38f8a1a9
 0869a438 00000000 000001a1
Aug 24 12:27:13 loxy kernel: 084cfc0c 060c1d48 28f99855 001507f6
0869a414
 02c4ff78 00000e5e 00000800
Aug 24 12:27:13 loxy kernel: 00000000 02c4ff7c 00000fff 060c1d00
0ffb0a40
 060c1d90 001368db 060c1d90
Aug 24 12:27:13 loxy kernel: Call Trace: [inet_recvmsg+114/136]
[sock_read+171/1
92] [net_bh+9/284] [sys_read+192/232] [system_call+85/124]
Aug 24 12:27:13 loxy kernel: Code: 8a 40 0d a8 02 74 04 ff 4c 24 10 8b 7c 24
10
39 7b 30 0f 87

~~~~~~

Walter Klomp,
Systems Manager / Administrator (Internet), Swiftech Automation Pte. Ltd.
   50 Kallang Avenue #09-01, Noel Corporate Building, Singapore 339505
  Tel. +65 391 8510, Fax. +65 274 4966, e-mail: walter@swiftech.net.sg.
==========================================================================
Swiftech Internet, the Value Added Service Provider with a personal touch.
==========================================================================

> -----Original Message-----
> From: Duane Wessels [mailto:wessels@ircache.net]
> Sent: Tuesday, August 24, 1999 7:17 AM
> To: Paolo Cova
> Cc: SQUID Users
> Subject: Re: cannot assign socket
>
>
>
>
> On Mon, 23 Aug 1999, Paolo Cova wrote:
>
> > I am running Squid Cache version 2.2.STABLE4 on a linux debian system
> > with kernel 2.2.8. Squid dies with this message in the log:
> >
> > 1999/08/23 20:25:37| WARNING: All dnsserver processes are busy.
> > 1999/08/23 20:25:37| WARNING: 1 pending requests queued
> > FATAL: Too many queued dnsserver requests
> > Squid Cache (Version 2.2.STABLE4): Terminated abnormally.
> >
> > What seems to be causing the problem is dnsserver, and below are the
> > repeated error messages in the log. I have looked for a solution on the
> > faq and the searchable archives, and read of similar problems from other
> > people, but found no solution so far.
> > Is squid trying to bind on the same port (0) of my X server or what?
> > This problem is independent of what port I assign squid in squid.conf
> > and of any tcp and udp incoming and outgoing addresses.
> >
> > 1999/08/23 20:25:37| helperOpenServers: Starting 5 'dnsserver' processes
> >
> > 1999/08/23 20:25:37| commBind: Cannot bind socket FD 1 to 127.0.0.1:0:
> > (99) Cannot assign requested address
> > 1999/08/23 20:25:37| commBind: Cannot bind socket FD 1 to 127.0.0.1:0:
> > (99) Cannot assign requested address
> > 1999/08/23 20:25:37| ipcCreate: Failed to create child FD.
> > 1999/08/23 20:25:37| WARNING: Cannot run
> > '/usr/local/squid/bin/dnsserver' process.
> >
> > Any insight?
>
> Does your linux system have a loopback interface configured?
>
> Can you ping 127.0.0.1?
>
> Duane W.
>
Received on Tue Aug 24 1999 - 12:13:12 MDT