Error From Stack - P_CONNECT_FAILED

[Girish Kurwalkar]

Hi,
I have OSI Stack running on HP-UX 11.00 Machine.

I am trying to communicate with this OSI stack from Solaris 2.8.

On the HP Side I am getting following error :

Error Received from stack:
P_CONNECT_FAILED

How do I get rid of this error. What this error means.

Any help will be really useful.

Thanks and Regards

Girish Kurwalkar

[T G Manikandan]

Are there any hpux error number with the P_CONNECT_FAILED message.

Any other messages.
Revert

[Girish Kurwalkar]

No Error Number in this message.

[Ron Kinner]

I would guess that the other end is not acking the original request from the HP for a connection. Perhaps it is not listening on the same port or whatever OSI calls it? Is there an equivalent tool to
netstat -a
that you can use to see that it is listening? Are you sure that the address of the Solaris matches what the HP is trying to reach? Does the Solaris know how to get to the HP's address?

Ron

[Michael Steele_2]

If the client is the HP then osidiag.

http://docs.hp.com/cgi-bin/fsearch/framedisplay?top=/hpux/onlinedocs/32070-90031/32070-90031_top.html&con=/hpux/onlinedocs/32070-90031/00/00/47-con.html&toc=/hpux/onlinedocs/32070-90031/00/00/47-toc.html&searchterms=stack%7cOSI&queryid=20030311-094325

[T G Manikandan]

First thing is that
tcp_bind: bad state, -4

indicates that you/application is trying to use a port which is already in use.

to find out which process is using that particular port you can use lsof to find that.


for the other message you can check this document
http://www2.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000063201322


Thanks

[T G Manikandan]

What is the OTS version you are running on HPUX.

PHNE_27889 is a patch for OTS for HPUX11.0
This patch requires a reboot

http://www2.itrc.hp.com/service/cki/patchDocDisplay.do?patchId=PHNE_27889

Do check it out

[Girish Kurwalkar]

Hi,
I am using OTS version C.09.02

[T G Manikandan]

Yes,that patch is for the version of OTS you have.

Do check whether the patch is already existent there.

Revert on the same

[Girish Kurwalkar]

Hi,
I have installed the patch. It was not there earlier.

Do I have to make any other configuration for this.

As I have started OTS and Checked my application, it still is giving me the same error.

[Michael Steele_2]

Are you using a version of BIND? Here are some rpc commands I'd like you to run, please run them before and after using the OSI stack and attach the results.

Thanks!

nfsstat -r (* display rpc information *)

rpcinfo -m (* rpcbind statistics*)

rpcinfo -p (* list of rpc programs *)

To show all of the RPC services on sapprd:

># rpcinfo sapprd

rstatd and rup (* rstatd daemon that supplies rup *)
rup client to server

To use uncomment these lines in /etc/inetd.conf

rpc xti tcp swait root /usr/dt/bin/rpc.ttdbserver 100083 1 /usr/dt/bin/rpc.ttdbserver
rpc dgram udp wait root /usr/dt/bin/rpc.cmsd 100068 2-5 rpc.cmsd

rup aruguements:
-h Sort the display alphabetically by host name.
-l Sort the display by load average.
-t Sort the display by up time.

[T G Manikandan]

What about the process which is using the port?
DID you find that out.Take off that process and try the same,

Thanks

[Girish Kurwalkar]

Hi TG,
Actually I missed that.

Just one question.

I have deleted the file nettl.LOG00 and rebooted the machine and restarted the OSI. I see feer tcp_bind: bad state, -4 errors. What does that mean. Is my OTS configuration is not done properly ?

[Girish Kurwalkar]

Hi,
I am attaching following Logs herewith.

nfsstat -r (* display rpc information *)

rpcinfo -m (* rpcbind statistics*)

rpcinfo -p (* list of rpc programs *)

[Girish Kurwalkar]

Hi,
Does anybody have an Idea about the following error :
***********************************HP OSI MGMT******************************@#%
Timestamp : Thu Mar 13 PST 2003 01:51:24.489513
Process ID : [ICS] Subsystem : OTS
User ID ( UID ) : -1 Log Class : ERROR
Device ID : -1 Path ID : -1
Connection ID : -1 Log Instance : 2
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[9600] High Access Method ERROR strosxtpi.c 2538
[....] (67) OsiTpiBindReq() address is busy.

***********************************HP OSI MGMT******************************@#%
Timestamp : Thu Mar 13 PST 2003 01:51:43.476315
Process ID : [ICS] Subsystem : OTS
User ID ( UID ) : -1 Log Class : ERROR
Device ID : -1 Path ID : -1
Connection ID : -1 Log Instance : 3
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[9600] High Access Method ERROR strosxtpi.c 2538
[....] (67) OsiTpiBindReq() address is busy.

[Michael Steele_2]

Are you getting any core files?

Do you see "Data Segmentation Fault" anywhere in dmesg or syslog.log?

Is Solaris 2.8 patched and configured OK?

We're exhausting the HP side.