Strange behavior in application

[Bryan_6]

I am in the process of planning/testing the migration of an application from a K580 to 7410. After restoring the data on the 7410, I started the app and discovered that there is a slight pause (about a second or so) when displaying any of the menus (it's a text-based application). I reviewed the kernel settings between servers and all appear consistent. I also watched Glance and validated that there were no resource constraints - CPUs are completely idle, memory is in abundance, etc. I thought it may be a network problem, but since I'm the only user currently, that shouldn't be an issue. What I'm speculating is that this is some sort of terminal problem as the delay/pause occurs at the same point in each menu. No other text-based interfaces (SAM, swinstall, etc.) experience this - it's just this particular application. Are there any system files I can look at that my shed some light on this problem? terminfo? What about kernel parms?

I realize this is a bit vague, but I tried to include as much information as possible... Thanks for the help.

[Bryan_6]

Another note:

During disaster recovery tests of the same data from the K580, I used Ignite to recover the OS on a 7410, and then used Omniback to recover the application data. However, in this situation (with the new server), the only difference from DR was that I left the OS as it was when HP pre-installed the 7410.

[Paul Sperry]

Try

export TERM=`ttytype -a`

and then try your program/app.

[A. Clay Stephenson]

There is not a lot to go on here.

1) Do an untic on both boxes and then diff them (obviously after ftp'ing to the same box).

2) Likewise do a stty -a on both boxes and diff them.

3) Do a kmtune on both boxes and diff them.

4) Are OS versions the same? Patch levels the same?

One thought is a speed/duplex mismatch between your new host and its switch port. Surprisingly, one end set to full and the other end set to half will do exactly this sort of thing and will be far from obvious in a telnet session which will appear to work normally.

[Dario_1]

Hi!

Check this out. Same problem you are having. I made some changes to the Kernel Parameters. Also, make sure the NIC card and switch are set to the same. The ideal will be:

100FD with auto-negotiation set to OFF.


Regards,

Dario

[Dario_1]

Hi again!

I forgot to add the post I opened. Here is it.

http://forums.itrc.hp.com/cm/QuestionAnswer/1,,0x4a185fe8b250d71190080090279cd0f9,00.html

Regards,

Dario

[Bryan_6]

All,

After a bit of troubleshooting using some of your suggestions, I discovered that this problem appears to be related a routing issue. I have the new server on a subnet different from my existing production server. Users on the same subnet work fine, while those on the other subnet have this slight delay. I've passed the issue along to my network guy and we're trying to figure out what's happening on the router.

Thanks for all of your help and suggestions!

Regards,
Bryan

[Bryan_6]

All,

After a bit of troubleshooting using some of your suggestions, I discovered that this problem appears to be related a routing issue. I have the new server on a subnet different from my existing production server. Users on the same subnet work fine, while those on the other subnet have this slight delay. I've passed the issue along to my network guy and we're trying to figure out what's happening on the router.

Thanks for all of your help and suggestions!

Regards,
Bryan