Re: [SQU] Squid 2.4 Head FATAL: Received Segment Violation...dying

Hi Henrik,

    I have changed the diskd storage for ufs, in order to get a backtrace,
and it crashes almost instantly, if you like me to clean my current cache in
order to make another test (or need a diferent test, I have a lot of
traffic, I can make it crash in just a few minutes, and have some results),
just let me know.

Here you have the backtrace:

Script started on Mon Feb 19 10:16:04 2001
GNU gdb 4.18
Copyright 1998 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain
conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB. Type "show warranty" for details.
This GDB was configured as "i386-unknown-freebsd"...
(gdb) handle SIGPIPE noprint nostop pass
Signal Stop Print Pass to program Description
SIGPIPE No No Yes Broken pipe
(gdb) r -CNd1
Starting program: /usr/local/sbin/squid -CNd1
2001/02/19 10:16:19| Starting Squid Cache version 2.4.PRE-STABLE for
i386-unknown-freebsd4.2...
2001/02/19 10:16:19| Process ID 205
2001/02/19 10:16:19| With 32768 file descriptors available
2001/02/19 10:16:19| Performing DNS Tests...
2001/02/19 10:16:19| Successful DNS name lookup tests...
2001/02/19 10:16:19| DNS Socket created on FD 7
2001/02/19 10:16:19| Adding nameserver 127.0.0.1 from /etc/resolv.conf
2001/02/19 10:16:19| Unlinkd pipe opened on FD 12
2001/02/19 10:16:19| Swap maxSize 43008000 KB, estimated 3308307 objects
2001/02/19 10:16:19| Target number of buckets: 165415
2001/02/19 10:16:19| Using 262144 Store buckets
2001/02/19 10:16:19| Max Mem size: 131072 KB
2001/02/19 10:16:19| Max Swap size: 43008000 KB
2001/02/19 10:16:19| Store logging disabled
2001/02/19 10:16:19| Rebuilding storage in /cache0/0 (DIRTY)
2001/02/19 10:16:19| Rebuilding storage in /cache1/0/0 (DIRTY)
2001/02/19 10:16:19| Rebuilding storage in /cache1/1/0 (DIRTY)
2001/02/19 10:16:19| Rebuilding storage in /cache1/2/0 (DIRTY)
2001/02/19 10:16:19| Rebuilding storage in /cache1/3/0 (DIRTY)
2001/02/19 10:16:19| Rebuilding storage in /cache1/4/0 (DIRTY)
2001/02/19 10:16:19| Rebuilding storage in /cache1/5/0 (DIRTY)
2001/02/19 10:16:19| Using Round Robin store dir selection
2001/02/19 10:16:19| Set Current Directory to /cache0/0
2001/02/19 10:16:19| Loaded Icons.
2001/02/19 10:16:19| Accepting HTTP connections at 0.0.0.0, port 80, FD 25.
2001/02/19 10:16:19| Accepting SNMP messages on port 1610, FD 26.
2001/02/19 10:16:19| Accepting WCCP messages on port 2048, FD 27.
2001/02/19 10:16:19| Ready to serve requests.
2001/02/19 10:16:23| Store rebuilding is 0.9% complete
2001/02/19 10:16:30| WARNING: newer swaplog entry for dirno 0, fileno
0000701E
2001/02/19 10:16:30| WARNING: newer swaplog entry for dirno 5, fileno
000004BC
2001/02/19 10:16:30| WARNING: newer swaplog entry for dirno 0, fileno
00007CE5
2001/02/19 10:16:31| WARNING: newer swaplog entry for dirno 5, fileno
000080D4

Program received signal SIGSEGV, Segmentation fault.
0x80b2416 in storeUfsDirReplRemove (e=0xbbecd00) at store_dir_ufs.c:1426
1426 SD->repl->Remove(SD->repl, e, &e->repl);
(gdb) bt
#0 0x80b2416 in storeUfsDirReplRemove (e=0xbbecd00) at store_dir_ufs.c:1426
#1 0x80afe2d in storeUfsDirRebuildFromSwapLog (data=0x9b9b400) at
store_dir_ufs.c:567
#2 0x806b456 in eventRun () at event.c:147
#3 0x80882ca in main (argc=2, argv=0xbfbffc24) at main.c:716
#4 0x804a529 in _start ()
(gdb) q
The program is running. Exit anyway? (y or n) y

If you think you need access to this server, I think I can provide it to
you, in order for you to debug this.

Greetings...
Ales

----- Original Message -----
From: "Henrik Nordstrom" <hno@hem.passagen.se>
To: "Jeremy Lyon" <jxlyon@uswest.com>
Cc: "squid users" <squid-users@ircache.net>
Sent: Sunday, February 18, 2001 2:45 PM
Subject: Re: [SQU] Squid 2.4 Head FATAL: Received Segment Violation...dying

> It would be very helpful if you could try to get a trace of where Squid
> fails for you without diskd (the only trace we have so far in this
> thread is a crash inside the diskd store).
>
> --
> Henrik Nordstrom
> Squid hacker
>
>
> Jeremy Lyon wrote:
> >
> > I have had this same problem not using diskd. That's why I haven't
> > pushed 2.4PRE-STABLE into production.
>
> --
> To unsubscribe, see http://www.squid-cache.org/mailing-lists.html

--
To unsubscribe, see http://www.squid-cache.org/mailing-lists.html