(Resending this with a subject and proper formatting - sorry).

We recently upgraded 2 BIND 9.4.2 servers running on Windows 2003 SP2 to
9.4.2-P1.  Both were running fine for a week or more.  However, one
failed at about 3:55 AM (EDT) yesterday..  The second failed about 7:23
AM.  Prior to the 9.4.2-P1 upgrade we had not experienced any problems
with either server.

These two servers are slaves to two Windows NT masters.  The load on
them is quite light (with some peaks if we get a flood of spam emails,
etc.).  The first failing server had been running for more than a week;
the second had been rebooted less than 15 hours before.  I saw no other
anomalous messages in any log prior to the crash.  The first failing
server runs a 64-bit version of Windows (rhea); the second runs a 32-bit
version (elara). The servers have nearly identical (but not exactly
identical) Windows patches.

Our query load is very low (2000 queries in about 90 minutes preceding
the first system crash).

Here is one of the crash event log messages:
    Faulting application named.exe, version 0.0.0.0, faulting module
ntdll.dll, version 5.2.3790.3959, fault address 0x00023705.

The second is similar (different fault address).

Here is our named.conf:

options {
            directory "D:/DNS";
            listen-on {
                        127.0.0.1;
                        192.76.186.12;
            };
            allow-query { any; };
            multi-master yes;
            check-names slave ignore;
            notify no;
            querylog no;
            statistics-file "etc/named.stats";
            dump-file "etc/named_dump.db";
            version "Unknown";
};

Zone "src.org" {
            type slave;
            masters {
                        192.76.186.3;
                        192.76.186.4;
            };
            file "Conf/named.src_org_int";
            zone-statistics yes;
};
 
zone "web.src.org" {
            type slave;
            masters { 192.168.2.110; };
            file "Conf/named.web_src_org_int";
            zone-statistics yes;
};

zone "186.76.192.in-addr.arpa" {
            type slave;
            masters {
                        192.76.186.3;
                        192.76.186.4;
            };
            file "Conf/named.192_76_186_int";
            zone-statistics yes;
};

zone "1.168.192.in-addr.arpa" {
            type slave;
            masters { 192.168.2.110; };
            file "Conf/named.192_168_1_int";
            zone-statistics yes;
};
 
zone "2.168.192.in-addr.arpa" {
            type slave;
            masters { 192.168.2.110; };
            file "Conf/named.192_168_2_int";
            zone-statistics yes;
};
 
zone "3.168.192.in-addr.arpa" {
            type slave;
            masters {
                        192.76.186.3;
                        192.76.186.4;
            };
            file "Conf/named.192_168_3_int";
            zone-statistics yes;
};
 
zone "13.27.216.in-addr.arpa" {
            type slave;
            masters {
                        192.76.186.3;
                        192.76.186.4;
            };
            file "Conf/named.216_27_13_int";
            zone-statistics yes;
};
 
zone "10.in-addr.arpa" {
            type slave;
            masters {
                        192.76.186.3;
                        192.76.186.4;
            };
            file "Conf/named.10_int";
            zone-statistics yes;
};
 
zone "localhost" {
            type master;
            file "Conf/localhost.zone";
};
 
zone "0.0.127.in-addr.arpa" {
            type master;
            file "Conf/0.0.127.in-addr.arpa.zone";
};
 
Have others experienced any problems with the patched Windows BIND
versions?
 
Let me know what else I can provide.

Any advice would be greatly appreciated.  I see the "debug" version of
9.4.2-P1 but no discussion on how to use it effectively (and these are
production servers).  Does 9.4.3b2 fix such a crash?
 
Thank you.
 
Bob Warren
Semiconductor Research Corp.
RTP, NC