[squid-users] ntlm auth fatal error

From: Steve Wilson Jr <SWilsonJr@dont-contact.us>
Date: Mon, 3 Jul 2006 21:03:18 -0400

I get this error quite a bit. It doesn't cause any real problems though.

[2006/07/03 13:13:13, 1] libsmb/ntlmssp.c:ntlmssp_update(267)
  got NTLMSSP command 3, expected 1

this is what worries me. 26 in queue but according to my reporting there
were only 22 people accessing the proxy between 1 and 2.

2006/07/03 13:17:24| WARNING: All ntlmauthenticator processes are busy.
2006/07/03 13:17:24| WARNING: up to 5 pending requests queued
2006/07/03 13:17:37| storeDirWriteCleanLogs: Starting...
2006/07/03 13:17:37| WARNING: Closing open FD 8
2006/07/03 13:17:37| Finished. Wrote 36113 entries.
2006/07/03 13:17:37| Took 0.0 seconds (1896093.7 entries/sec).
FATAL: Too many queued ntlmauthenticator requests (26 on 5)
Squid Cache (Version 2.5.STABLE14): Terminated abnormally.
CPU Usage: 777.645 seconds = 435.479 user + 342.165 sys
Maximum Resident Size: 0 KB
Page faults with physical i/o: 0
Memory usage for squid via mallinfo():
        total space in arena: 24064 KB
        Ordinary blocks: 23338 KB 23 blks
        Small blocks: 0 KB 0 blks
        Holding blocks: 1240 KB 2 blks
        Free Small blocks: 0 KB
        Free Ordinary blocks: 725 KB
        Total in use: 24578 KB 102%
        Total free: 725 KB 3%

I could add more helpers but would this really solve the problem? We
plan on putting over 300 people on this proxy in a week and a half. This
has happened twice and squid has always restarted in 3 seconds. A 3
second hiccup multiple times in a day would be a problem. The access log
doesn't show anything helpful and there's nothing in squid.out. Any help
would be appreciated. Thank you.

Steve Wilson Jr
Loxias IT Solutions
513-605-2726
swilsonjr@loxias.com
Received on Mon Jul 03 2006 - 19:03:31 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Aug 01 2006 - 12:00:01 MDT