squid 1.1.3: "ipcache_dnsHandleRead: bad status"

We're running squid 1.1.3 on a DEC AlphaStation (DEC OSF/1 3.0b).
Last night, squid crashed, with the following entry in the cache.log:

97/01/09 02:56:32| ERR_DNS_FAIL:http://www.hqs.com/hqs/linkexchange/hqs-ba05.gif
FATAL:ipcache_dnsHandleRead: bad status
Squid Cache (Version 1.1.3):Terminated abnormally.
CPU Usage: user 3206 sys 2257
Memory Usage: rss 24960 KB
Page faults with physical i/o: 187
Memory usage for squid via mallinfo():
        total space in arena: 23146 KB
        Ordinary blocks: 7829 KB 53805 blks
        Small blocks: 8190 KB 104500 blks
        Holding blocks: 0 KB 1045 blks
        Free Small blocks: 1834 KB
        Free Ordinary blocks: 5291 KB
        Total in use: 16019 KB 69%
        Total free: 7126 KB 31%
97/01/09 02:56:39| storeWriteCleanLog: Starting...

...and the usual "x lines written so far" messages that happen during
shutdown.

Fortunately, this happened in the middle of the night - not exactly
"peak" time. :=] Any clues?

FYI: DEC OSF/1 3.0b, built with gcc (no gnumalloc, no extra compile
flags), 6 dnsserver processes at squid startup, cache_mem 8 MB,
cache_swap 700 MB, cache_mem_low 40, cache_mem_high 60. If there is
some other relevant snippet of information that I haven't included,
please let me know.

Thanks for any and all responses.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
= Kirk Friggstad = Western Computer Link =
= kirk@the.link.ca = webmaster@link.ca =
= http://www.link.ca/~kirk = http://www.link.ca =
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Received on Thu Jan 09 1997 - 08:00:53 MST