Hello all.

Here in productive work is BM37 working on NW60.
[P4, 512RAM, Adaptec 29169, 2 x IntelPro100 (driver ce100b from
NW6SP3)]
Until recently were it is established NW60SP2+Postfixes and
BM37+SP1+POSTFIXES.

After installation of these updates on NW6.0:
nw6sp3

And postfixes:
tcp607l
nw6nss3c
NICI 2.6.0
flsysft9
mm6up1c
nwlib5a
tsa5up13
wsock6c
nls603ft
httpstk3
nw56up3
ncmft1
nwmem3
ndb410x
edir862sp5
secupd1.tgz
jvm131sp5
For NetWare 6

And Updates for BM 3.7:
b4bm37sp3

In work Stateful TCP Filter Exceptions strange failures began to arise
-
after intensive normal work in current some hours filters
spontaneously
break off connections, at attempt to retry connection it is restored
for
short time then traffic again blocked.

If from the console to execute next commands:

unload ipflt
unload ipxflt
unload filtsrv

load ipflt
load ipxflt
load filtsrv

that work of filters is restored for some long time.

On the sever console no any error messages. Modules of system of a
packet
filtration are unloaded as without errors.

Turning back of modules from BM37SP2+bm37fp3f [filtsrv.nlm +
ipflt31.nlm
+ipflt.nlm] has not removed the given problem.

It is possible to carry out certainly NCF on reloading filters from
CRON.NLM
twice day:), but it not solution. Application of modules from
structure
b4bm37sp3.exe also does not correct a state of affairs.

Whether there are at somebody ideas where and what to change?

Dmitry E. Rovkin