Solaris 7 BIND core dumping

[Josh McKee]

Recently our Solaris 7 name servers began dumping core. This has
happened 4 times over the last 4 weeks. The message logged in syslog
is:

Sep 27 07:13:35 gateway named[19620]: ../common/ns_main.c:517:
INSIST(evGetNext(ev, &event, EV_WAIT) != -1): Not enough space failed.

Can anyone explain this error message to me? I have configured
resource limites as follows:

coresize        128M;
datasize        64M;
stacksize       16M;
files           128;

The affected systems are both running the stock Solaris 7 name server
(8.1.2). One system has 512MB of memory, the other 256MB. Disk space
looks good (tmp is at 1% capacity). syslog does report any out of disk
space messages around the times the name server has crashed, and
everything else is running without problem.

Any help would be greatly appreciated.

Josh