Re: TCP goes down

[Wayne McAlpine <wayne.mcalpine@xxxxxxxxxxxxxxxxx>]

I agree with Rich.  If the ethernet link hiccups, an inquiry message is 
generated once the recovery limits are reached.  The system default 
value QCMNRCYLMT is 2 and 5, which means that the system will make two 
attempts at recovery, 5 minutes apart, before sending an inquiry message 
CPA58EE to the system operator message queue.

You can specify larger recovery limits on the individual ethernet lines 
and you can also add the inquiry message to the system reply list so 
that the recovery can proceed completely unattended.



Rich Loeber wrote:
> Sheri,
>
> I've seen this happen when the cable connection is disturbed or the router (or
> hub) it is connected to looses power for a very short time.  The first place
> I've learned to look is at the QSYSOPR message queue.  If there is an error
> there, then a simple 'R' to retry gets you back on track (assuming that the
> physical problem has been corrected).
>
> Rich Loeber
> Kisco Information Systems
> http://www.kisco.com
>   ------------------------------------------------------------------------
>
> "Rowe, Sheri" wrote:
>
> > Has anyone had a problem with their TCP connection 'dropping'?  About 8
> > months ago, our overnight failed as the TCP on one of our iSeries went
> > 'down'.  The system was up and I could sign on to the console fine but
> > not from a Client Access session.  Tried stopping and starting TCP (all
> > services) and varying off and on the line, but only an IPL could get the
> > communication between the iSeries and my network up.
> >
> > I contacted IBM who couldn't really help.  No entries in WRKPRB either.
> > I couldn't ping anything.
> >
> > The history log had a bunch of entries to indicate the interactive
> > sessions and the connection to our other iSeries had terminated
> > TCP2617 TCP/IP connection to remote system 169.1.1.25 closed, reason
> > code 2. TCP2617 TCP/IP connection to remote system 169.1.1.25 closed,
> > reason code 3. CPF590A Session to device WT221S2 ended normally.
> >
> > It happened again this past Friday.  Does anyone know what is causing
> > this?  And/or how I can correct it once it does happen without an IPL?
> >
> > Thanks for any words of wisdom.
> >
> > Sheri Rowe
> > Timex Canada
> > Markham, Ontario
> >
> > --
> > This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
> > To post a message email: MIDRANGE-L@xxxxxxxxxxxx
> > To subscribe, unsubscribe, or change list options,
> > visit: http://lists.midrange.com/mailman/listinfo/midrange-l
> > or email: MIDRANGE-L-request@xxxxxxxxxxxx
> > Before posting, please take a moment to review the archives
> > at http://archive.midrange.com/midrange-l.