In my view this indicates some kind of socket problem, very possibly an AIX
problem. The popup is generated when nvcorrd does not respond to a
request. Most often this means that he's down. But the error message
indicates that he was denied a socket connection for some reason. That's
all I can tell you.
I doubt you will get a solution here, unless someone can recommend AIX
maintenance. I posted this awhile back:
>>The following AIX PTFs have been PE'd:
AIX 4.2.1 PTF: bos.net.tcp.client 4.2.1.22 (IX84982)
AIX 4.3.2 PTF: bos.net.tcp.client 4.3.2.4 (IX85539)
>> If you get a customer on AIX 4.2.1 or 4.3 with strange socket issues,
e.g. netstat -an shows massive queue backups >>without an obvious reason,
check the lslpp -ha for one of these filesets. The AIX 4.2.1 PTF has been
replaced by >>bos.net.tcp.client 4.2.1.23. Rejecting the fileset is
currently the only 4.3 solution. This bug (IX87310) hoses the socket when
>>large data transfers occur over large MTU interfaces, i.e. loopback.
Problems reported, so far, include >>nvcorrd(1666)->nvserverd(1665) socket
problems, often masquerading as ruleset problems.
Other folks said here that there were later versions of this code
available.
So perhaps you should check with AIX and make sure you have the latest
maint for socket problems.
James Shanks
Tivoli (NetView for UNIX) L3 Support
Matthias Widmann <mwidmann@DEBIS.DE> on 05/20/99 12:41:50 PM
Please respond to Discussion of IBM NetView and POLYCENTER Manager on
NetView <NV-L@UCSBVM.UCSB.EDU>
To: NV-L@UCSBVM.UCSB.EDU
cc: (bcc: James Shanks/Tivoli Systems)
Subject: Re: nvevents failure
Hi All,
we've got the same problem,
NvEvents stops (YES it run before) working showing a popup that complains
about
connection loss to nvcorrd.
Contrary to my expectations nvcorrd was running!
The popup also showed that NvEvents will reconnect to nvcorrd, so it did -
with a loss of events -> so some correlation didn't work very well then.
Our installation is:
AIX 4.2.1
NV 5.1.1
nv6000.log also shows:
Acceptor:: Address already in use.
ulimits and max processes are set properly
Does anybody else has the same problem or better a solution?
I hope someone has - Sorry - a solution!
Best regards
Matthias Widmann
James Shanks wrote:
> I have never seen ANYTHING like this before. Do other people telnet
into
> this box and export the display back to run NetView? If so, do a ps -ef
> and see how many there are. It sounds as if nvevents cannot open a
socket
> (presumably to nvserverd but I am guessing). Did you follow the advice
in
> the Release Notes (pages 9 and 10) about increasing the max processes per
> user and adding the ulimit statements in etc/rc.nfs. If you are root,
> try issuing them from the command line and see if that helps. If you are
> not root, try it as root.
>
> Otherwise, what I would do is go back to a known, minimal state and see
> what happens. Reboot if possible.
> If you get this same error when no other GUIs are running, then please
call
> Support. That's all you can do.
>
> James Shanks
> Tivoli (NetView for UNIX) L3 Support
>
> HVILLEGA <hvillega@BANCOLOMBIA.COM.CO> on 05/19/99 09:51:35 AM
>
> Please respond to Discussion of IBM NetView and POLYCENTER Manager on
> NetView <NV-L@UCSBVM.UCSB.EDU>
>
> To: NV-L@UCSBVM.UCSB.EDU
> cc: (bcc: James Shanks/Tivoli Systems)
> Subject: nvevents failure
>
> Hi All:
> I have somehing wrong with the nvevents application, when I run the
> NetView the
> events window does not start , and in the nv6000.log arrives the
following
> message:
> Acceptor:: Address already in use.
>
> I have tried to start the nvevents manually but it does not start.
>
> Does anybody know what is wrong ?
>
> My enviroment is NV5.1.1 aix 4.3.2.
>
> Thanks and regards,
>
> Hernan D. Villegas.
|