Or the auto-kill/kill-all could check whether instances are killed after some time, and try again if not... Another option I see right now is to reboot the server once per day and use a scheduled task to launch all of this (I have it already scripted). The one thing I would have to figure out is how to click the StartAll, but I think there may be a command line option... But rebooting the server is a little brute-force...

Thanks
Andy
KU7T

From: Ted Gisske <gisske@offex.com>
Sent: Friday, July 26, 2019 9:53 AM
To: 'Andy KU7T' <ku7t@ku7t.org>; 'W3OA' <w3oa@roadrunner.com>; skimmertalk@contesting.com <skimmertalk@contesting.com>
Subject: RE: [Skimmertalk] FT8StartUp Beta Version 2.2b1
 

Andy,

I’ve noticed the same thing. Some instances of WSJT-X  and CWSL_SSBWave.exe seemed slow to close on a kill rigs command. The, if I hit the start-all button before all instanced died, there were a lot of warnings about duplicate instances already running.

 

As a suggestion…Perhaps 15-20 seconds of dawdle time before auto-starting the rigs after an auto-kill would improve the issue.

 

73,

Ted

K9IMM

 

From: Skimmertalk [mailto:skimmertalk-bounces@contesting.com] On Behalf Of Andy KU7T
Sent: Friday, July 26, 2019 11:42 AM
To: W3OA; skimmertalk@contesting.com
Subject: Re: [Skimmertalk] FT8StartUp Beta Version 2.2b1

 

After running with the kill switch for a week, I noticed that my spot count went way down (1100 for the last 24 hours). I hit the kill all switch and found this:

  • all UI instances of WSJT-X were gone - good
  • 24 CWSL_SSBWave.exe instances were still running - not good
  • 2 wsjt-x.exe instance were running - not good

I killed the 26 orphan processes and hit the start all switch. Things are now moving again. 

 

2 Questions:

  • Can the kill all switch be a little more aggressive and do full kill of these processes?
  • Do you think that a high CPU situation can cause fewer spots?  I am wondering whether 80 -90 CPU utilization may throttle already the spot count. I wonder if running on fewer bands would be better. Anyone know or have tried this?

Thanks

Andy

KU7T

From: Andy KU7T
Sent: Sunday, July 14, 2019 8:30 AM
To: W3OA <w3oa@roadrunner.com>; skimmertalk@contesting.com <skimmertalk@contesting.com>
Subject: RE: [Skimmertalk] FT8StartUp Beta Version 2.2b1

 

Dick,

 

I ran for a few days with the “automatic kill switch”. I am running the new RP on 8 FT8 bands. It seems to work for the WSJT-X and jt9 processes. However, I can see currently 16 processes with names CWSL_SSBWave.exe. Anything you can do here? I know I could use a script also to do that easily, but it would interfere with the FT8Startup functionality, so it would be better if it would do it….

 

Thanks

Andy

KU7T

 

Sent from Mail for Windows 10

 

From: Skimmertalk <skimmertalk-bounces@contesting.com> on behalf of W3OA <w3oa@roadrunner.com>
Sent: Thursday, June 27, 2019 5:11:42 PM
To: skimmertalk@contesting.com
Subject: [Skimmertalk] FT8StartUp Beta Version 2.2b1

 

FT8StartUp Beta Version 2.2b1 is available for testing at
https://www.dropbox.com/s/f3rjjuzra51364v/FT8StartUp.exe?dl=0

This adds the option to "Kill All Rigs" and then "Start All Rigs in Use"
at 2:00 am every day,   This is a work around for the problem that
causes over driving the audio input to WSJT-X after long periods of
operation with CWSL_SSBWave.exe.

73 - Dick, W3OA
_______________________________________________
Skimmertalk mailing list
Skimmertalk@contesting.com
http://lists.contesting.com/mailman/listinfo/skimmertalk