Can you find anything in the logs for dlsd?
Am 09/09/2013 08:04 PM, schrieb Johnathan Van Why:
On Mon, Sep 9, 2013 at 7:39 AM, Richard Hacker <[email protected]> wrote:
Are you using a multi-processor system? If so, can try using HPET (although
this timing system is buggy by design - was developed together with Microsoft
;) )
I switched to the HPET, and it still drops data.
Thank You,
Johnathan Van Why
Dynamic Robotics Laboratory
Oregon State University
Am 09/06/2013 09:25 PM, schrieb Johnathan Van Why:
On Thu, Sep 5, 2013 at 11:38 PM, Richard Hacker <[email protected]> wrote:
What is you underlying real-time operating system?
I am doing my testing on Linux, patched with PREEMPT_RT for realtime.
The DLS has a problem when the RT system influences time on the PC. This has
been experienced with RTAI.
Ways to fix it? Watch the system's clock while the control system is loaded
and make sure it is not "pulled" in any way.
I have confirmed that there is no clock drift (note: the kernel's
clocksource is the TSC). I disabled NTP and re-ran the test, with the
same results (lost data and no clock drift).
Do you have any other ideas?
Thank you for the help,
Johnathan Van Why
Dynamic Robotics Laboratory
Oregon State University
Am 09/06/2013 04:33 AM, schrieb Johnathan Van Why:
We have been benchmarking EtherLab (we're considering switching to it
as our robot control system), and have noticed that the Data Logging
Service has been dropping data. We created over 100 independent
sources in a Simulink model, deployed to EtherLab, and logged all
sources at 1 kHz (over 400 KB/s total).
After looking at the log data, we noticed that a significant amount of
data was dropped. Every 15 minutes, the time variable would jump by 3
minutes; the total data we collected was significantly less than we
expected for the duration of the run.
Is this a known issue? Are we asking too much of the Data Logging
Service? Is there anything we can try in order to obtain reliable
logging at this data rate?
Thank you for any help,
Johnathan Van Why
Dynamic Robotics Laboratory
Oregon State University
_______________________________________________
etherlab-users mailing list
[email protected]
http://lists.etherlab.org/mailman/listinfo/etherlab-users
Mit freundlichem Gruß
Richard Hacker
--
------------------------------------------------------------------------
Richard Hacker M.Sc.
[email protected]
Tel.: +49 201 / 36014-16
Ingenieurgemeinschaft IgH
Gesellschaft für Ingenieurleistungen mbH
Heinz-Bäcker-Str. 34
D-45356 Essen
Amtsgericht Essen HRB 11500
USt-Id.-Nr.: DE 174 626 722
Geschäftsführung:
- Dr.-Ing. T. Finke,
- Dr.-Ing. W. Hagemeister
Tel.: +49 201 / 360-14-0
http://www.igh-essen.com
------------------------------------------------------------------------
Mit freundlichem Gruß
Richard Hacker
--
------------------------------------------------------------------------
Richard Hacker M.Sc.
[email protected]
Tel.: +49 201 / 36014-16
Ingenieurgemeinschaft IgH
Gesellschaft für Ingenieurleistungen mbH
Heinz-Bäcker-Str. 34
D-45356 Essen
Amtsgericht Essen HRB 11500
USt-Id.-Nr.: DE 174 626 722
Geschäftsführung:
- Dr.-Ing. T. Finke,
- Dr.-Ing. W. Hagemeister
Tel.: +49 201 / 360-14-0
http://www.igh-essen.com
------------------------------------------------------------------------
Mit freundlichem Gruß
Richard Hacker
--
------------------------------------------------------------------------
Richard Hacker M.Sc.
[email protected]
Tel.: +49 201 / 36014-16
Ingenieurgemeinschaft IgH
Gesellschaft für Ingenieurleistungen mbH
Heinz-Bäcker-Str. 34
D-45356 Essen
Amtsgericht Essen HRB 11500
USt-Id.-Nr.: DE 174 626 722
Geschäftsführung:
- Dr.-Ing. T. Finke,
- Dr.-Ing. W. Hagemeister
Tel.: +49 201 / 360-14-0
http://www.igh-essen.com
------------------------------------------------------------------------
_______________________________________________
etherlab-users mailing list
[email protected]
http://lists.etherlab.org/mailman/listinfo/etherlab-users
