In article <6BB570CE6EBAD0118B6300805FFEE2CFC1CF1D AT LMIG-MSG-01.lmig DOT com>,
"Robinson, Cris" <Cris.Robinson AT LIBERTYMUTUAL DOT COM> says:
>
>I also get this message.  I noticed it happening about the time we
>updated the server with ptf 3 on our NT server. I have 300MB of memory
>on my server with about 400 clients. The only way I can find to clear it
>is to reboot the server.  Sounds fishy to me.
>
>CR
>
We also have seen this problem. Have you reported this to IBM?

We have an open problem with IBM, but they have not found the problem
yet. IBM has asked us to run a server trace with systime and tcp
enabled at the time the problem occurs. But we have not seen the
problem since we started doing the following:

Halting and restarting the adsm server will also clear the problem.
And you can do this before the problem occurs, preventing the problem.

It appears that wnt windows nt server can build up quite a number
of listen socket connections. Try a netstat -a command on the
windows nt machine to see the open socket connections.
We found that over time, the number of socket listens would
keep increasing. If we halted the adsm server, without changing
anything else, most of the listen socket connection would go away.
I suspect that these are related to node(s) that the adsm scheduler
prompter is having problems with.

Please let us know if you find out anything more -- len

-------------------------------------------------------------------------
Leonard Boyle, Mainframe support            snolen AT vm.sas DOT com
Leonard Boyle, Mainframe support            snolen AT vm.sas DOT com
SAS Institute Inc.                          ussas4hs@ibmmail
Room E206                                   (919) 677-8000 ext 6241
203 SAS Campus Drive
Cary NC 27513