[squid-users] HEEEELLLPPPP

From: Chris Knipe <savage@dont-contact.us>
Date: Sun, 25 May 2003 00:54:48 +0200

Lo all,

My squid all of a sudden after a reboot stopped working. I have *NO* idea
what is causing this, but it is behaving TOTALLY strange. Starting squid in
full debugging (-X) for example, logs NO debugging information. Telling
squid not to start a daemon (-N) for example, and it still starts a
daemon.... Hence, it's broken.

The only thing that I was able to pick up, was a single error in the
cache.log, that was never there before....

2003/05/25 00:47:08| Starting Squid Cache version 2.4.STABLE7 for
i386-portbld-freebsd4.8...
2003/05/25 00:47:08| Process ID 16337
2003/05/25 00:47:08| With 14781 file descriptors available
2003/05/25 00:47:08| Performing DNS Tests...
2003/05/25 00:47:08| Successful DNS name lookup tests...
2003/05/25 00:47:08| DNS Socket created on FD 4
2003/05/25 00:47:08| Adding nameserver 192.168.1.1 from /etc/resolv.conf
2003/05/25 00:47:08| Unlinkd pipe opened on FD 9
2003/05/25 00:47:08| Swap maxSize 262144 KB, estimated 26214 objects
2003/05/25 00:47:08| Target number of buckets: 1310
2003/05/25 00:47:08| Using 8192 Store buckets
2003/05/25 00:47:08| Max Mem size: 16384 KB
2003/05/25 00:47:08| Max Swap size: 262144 KB
2003/05/25 00:47:08| Rebuilding storage in /var/spool/squid (DIRTY)
2003/05/25 00:47:08| Using Least Load store dir selection
2003/05/25 00:47:08| Set Current Directory to /var/spool/squid
2003/05/25 00:47:08| assertion failed: internal.c:99: "host && port && name"
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2003/05/25 00:47:08| Starting Squid Cache version 2.4.STABLE7 for
i386-portbld-freebsd4.8...
2003/05/25 00:47:08| Process ID 16337
2003/05/25 00:47:08| With 14781 file descriptors available
2003/05/25 00:47:08| Performing DNS Tests...
2003/05/25 00:47:08| Successful DNS name lookup tests...
2003/05/25 00:47:08| DNS Socket created on FD 4
2003/05/25 00:47:08| Adding nameserver 192.168.1.1 from /etc/resolv.conf
2003/05/25 00:47:08| Unlinkd pipe opened on FD 9
2003/05/25 00:47:08| Swap maxSize 262144 KB, estimated 26214 objects
2003/05/25 00:47:08| Target number of buckets: 1310
2003/05/25 00:47:08| Using 8192 Store buckets
2003/05/25 00:47:08| Max Mem size: 16384 KB
2003/05/25 00:47:08| Max Swap size: 262144 KB
2003/05/25 00:47:08| Rebuilding storage in /var/spool/squid (DIRTY)
2003/05/25 00:47:08| Using Least Load store dir selection
2003/05/25 00:47:08| Set Current Directory to /var/spool/squid
2003/05/25 00:47:08| assertion failed: internal.c:99: "host && port && name"
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

After that, the squid process dies, and a new process is started. At the
end of the day, I sit with 20+ squid processes running (some accepting
connections), but my system's load sky rockets to well over 30!!!!
Eventually, when I do manage to get a single squid process running and
accepting connections, it consumes on avarage like between 80% and 90% of
both memory as well as CPU resources...

What gives... I'd *love* to give more debugging information... But alas, the
first paragraph said it all... :(

--
me
Received on Sat May 24 2003 - 16:55:21 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:17:01 MST