[squid-users] squid 2.5 - ipf transparent proxy - FreeBSD 5.3-p13

Hello Squid users,

I upgraded my Squid 2.5 STABLE9 to STABLE10, I used the
--enable-ipf-transparent make arg for transparent proxying with ipf
(version 3.4.35 on FreeBSD 5.3-p13 system).

I used the same squid.conf which I have been using in STABLE9 (without
any problems). Starting squid STABLE10 is no problem. No errors in
cache.log, with squid -k parse also no problems.

But when a client starts browsing, the squid process is exited with
signal 6 and a core dump. This is a fragment of /var/log/messages:

May 21 11:28:26 appelstroop kernel: pid 699 (squid), uid 100: exited on
signal 6 (core dumped)

This is what the cache.log of squid says:

2005/05/21 11:28:29| Starting Squid Cache version 2.5.STABLE10 for
i386-portbld-freebsd5.3...
2005/05/21 11:28:29| Process ID 720
2005/05/21 11:28:29| With 3520 file descriptors available
2005/05/21 11:28:29| DNS Socket created at 0.0.0.0, port 60013, FD 5
2005/05/21 11:28:29| Adding nameserver 217.194.109.15 from squid.conf
2005/05/21 11:28:29| Adding nameserver 217.194.96.10 from squid.conf
2005/05/21 11:28:29| Adding nameserver 217.194.97.13 from squid.conf
2005/05/21 11:28:29| Unlinkd pipe opened on FD 10
2005/05/21 11:28:29| Swap maxSize 2560000 KB, estimated 196923 objects
2005/05/21 11:28:29| Target number of buckets: 9846
2005/05/21 11:28:29| Using 16384 Store buckets
2005/05/21 11:28:29| Max Mem size: 8192 KB
2005/05/21 11:28:29| Max Swap size: 2560000 KB
2005/05/21 11:28:29| Rebuilding storage in /cache (CLEAN)
2005/05/21 11:28:29| Using Least Load store dir selection
2005/05/21 11:28:29| Set Current Directory to /cache
2005/05/21 11:28:29| Loaded Icons.
2005/05/21 11:28:29| Accepting HTTP connections at 0.0.0.0, port 8080,
FD 12.
2005/05/21 11:28:29| Accepting ICP messages at 0.0.0.0, port 3130, FD
13.
2005/05/21 11:28:29| WCCP Disabled.
2005/05/21 11:28:29| Ready to serve requests.
2005/05/21 11:28:29| Done reading /cache swaplog (46 entries)
2005/05/21 11:28:29| Finished rebuilding storage from disk.
2005/05/21 11:28:29| 46 Entries scanned
2005/05/21 11:28:29| 0 Invalid entries.
2005/05/21 11:28:29| 0 With invalid flags.
2005/05/21 11:28:29| 46 Objects loaded.
2005/05/21 11:28:29| 0 Objects expired.
2005/05/21 11:28:29| 0 Objects cancelled.
2005/05/21 11:28:29| 0 Duplicate URLs purged.
2005/05/21 11:28:29| 0 Swapfile clashes avoided.
2005/05/21 11:28:29| Took 0.3 seconds ( 158.7 objects/sec).
2005/05/21 11:28:29| Beginning Validation Procedure
2005/05/21 11:28:29| Completed Validation Procedure
2005/05/21 11:28:29| Validated 46 Entries
2005/05/21 11:28:29| store_swap_size = 242k
2005/05/21 11:28:30| storeLateRelease: released 0 objects

No problems at all, as you can see.

I already recreated the cache files/dirs with squid -z, I checked the
permissions, they are OK.

Does somebody know how to solve this problem?

Best regards,
Martijn Broeders
m.broeders@hublabs.nl
Received on Mon May 23 2005 - 03:59:03 MDT