Testing tool causes 100% CPU utilization

[Jerry Tilsley]

All,

I’m running CL 5.8.4 on RedHat, and am having an issue with the Testing Tool.  When I am performing a TPS test (this happens with CL 5.8.0 as well) the CPU utilization spikes to 100% and hangs, no output is ever generated.  There are no issues with these scripts if I actually use them in a thread, only when running through the testing tool.  Also, the problem does not happen if the data file is only a few hundred lines, but more than that and the tool hangs and spikes the CPU.  Any suggestions out there?

Thanks!

Topic

[Donna Hooe]

Our organization has 2 Cloverleaf Linux boxes, one is local and the other is remote hosted.  I have seen the problem you describe on my remote hosted box but everything works fine on the local box.  We are running 5.8.4 on the remote box and 5.8.3 on the local box but we had the issue on the remote box with 5.7.3 as well.

In our case we suspect that all the firewall security is the problem, since we don’t see the problem with our local box.   I have a call in about this but no one has been able to come up with an explaination or a fix.  If I hear back from someone about this and we find a solution I will update you.

[Jerry Tilsley]

My situation is similar but different.  Both the server and the client are local, but both are virtual machines.  I will have them look to see if they are on separate virtual switches as well.  Thanks for information!

[Donna Hooe]

Both of our Linux boxes are also virtual.

[Jerry Tilsley]

I just ran the command from the local console on the server, still the same issue.  I do not believe this is network related.  Something going on with the server itself I suspect, just haven’t been able to pinpoint it.

[Donna Bailey]

Jerry,

Did you ever get a resolution to this…I spent hours last night trying to test a change I need to make and the GUI wouldn’t work or the command line.  I’m getting to the windows cloverleaf server using remote desktop and I had to kill off all the hcixlttest.exe’s to bring the CPU down.  

Just curious if you figured out what was going on???

Thanks,

Donna

Donna Bailey
Tele: 315-729-3805
dbailey@microstar.health
Micro Star Inc.

[Jerry Tilsley]

It had something to do with the encoding that was selected.  Cannot remember which one, what are you using for the encoding?  ASCII?

[Donna Bailey]

ASCII defaults and I don’t change it….

Donna

Donna Bailey
Tele: 315-729-3805
dbailey@microstar.health
Micro Star Inc.

[Jerry Tilsley]

Try changing it to blank (none), I believe this is what support had be do to work around the issue.  I believe this issue is resolved with CL 5.8.5 which is what I am on now.

[Donna Bailey]

Isn’t that the oddest thing….it worked!!!  I know I’ve used the testing tool in the past….hmmmm

Thank You so much for your help!!

Donna

Donna Bailey
Tele: 315-729-3805
dbailey@microstar.health
Micro Star Inc.

[Donna Bailey]

I think I figured out another part of the problem…the message I was testing with had over 900 OBX’s….

Thanks again for your help!

Donna

Donna Bailey
Tele: 315-729-3805
dbailey@microstar.health
Micro Star Inc.

[Paul Johnston]

Thankyou Jerry and Donna,

I too have run into this issue a couple of times. Once with a large amount of OBX segments in the message and just recently with a CSV file with over 6000 records.

I was able to successfully Test both scenarios with the ENCODING option set to Blank .

[Author]

Replies