nci not starting asap as expected


Advanced search

Message boards : Number crunching : nci not starting asap as expected

Author Message
Profile Saenger
Send message
Joined: 16 Jun 11
Posts: 166
Credit: 635,544
RAC: 102

Message 24 - Posted: 17 Jun 2011 | 1:44:54 UTC
Last modified: 17 Jun 2011 | 1:45:20 UTC

I've just attached the project, got 4 WUs (why isn't there a limit to 1 per computer) but none of those started asap, although both are at least named "nci".
I don't know whether there is anything else to do for BOINC to recognize nci-WUs, do they have to start once? Is there a special software marker?

I get it, that all of those 4 will run in an error as long as I don't have the external devise, right? So I will see whether they are gone once I come home after work ;)
____________
Gruesse vom Saenger

Profile krzyszp
Project administrator
Project developer
Project tester
Project scientist
Avatar
Send message
Joined: 16 Apr 11
Posts: 383
Credit: 787,492
RAC: 102

Message 25 - Posted: 17 Jun 2011 | 3:46:31 UTC - in response to Message 24.

We still work on app and server configuration.

All your tasks come unfortunately with error, because we work on app permissions on Linux systems. Basic problem is that application require special hardware but manager on default linux configuration doesn't have enough permission to access USB port...

Anyway, thx for info
____________
Regards,
Krzysztof 'krzyszp' Piszczek
Android Radioactive@Home Map
Android Radioactive@Home Map - donated
My Workplace

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,382,673
RAC: 45

Message 27 - Posted: 17 Jun 2011 | 16:36:31 UTC - in response to Message 24.

I've just attached the project, got 4 WUs (why isn't there a limit to 1 per computer) but none of those started asap, although both are at least named "nci".
I don't know whether there is anything else to do for BOINC to recognize nci-WUs, do they have to start once? Is there a special software marker?


The server was slightly misconfigured. Right now it handles the work as "nci".
The app has been changed a bit, instead of exiting immediately if the sensor is not present, it waits a while.

Profile Saenger
Send message
Joined: 16 Jun 11
Posts: 166
Credit: 635,544
RAC: 102

Message 28 - Posted: 17 Jun 2011 | 16:48:07 UTC

This time it started asap as ecpected, it was only one DLed, it runs as nci with 0.01 CPU, it doesn't error out.
Looks fine :D
____________
Gruesse vom Saenger

Profile Saenger
Send message
Joined: 16 Jun 11
Posts: 166
Credit: 635,544
RAC: 102

Message 29 - Posted: 18 Jun 2011 | 3:43:24 UTC

They error with this one after nearly exact 6000sec:

<core_client_version>6.10.58</core_client_version>
<![CDATA[
<message>
process exited with code 1 (0x1, -255)
</message>
<stderr_txt>
error finding DataStore: The specified device was not found

</stderr_txt>
]]>

As they don't use any ressources it's fine with me that I don't get any reward, but are you testing something with them or just the server?
____________
Gruesse vom Saenger

Profile TJM
Project administrator
Project developer
Project tester
Send message
Joined: 16 Apr 11
Posts: 291
Credit: 1,382,673
RAC: 45

Message 30 - Posted: 18 Jun 2011 | 8:10:40 UTC - in response to Message 29.

These workunits are used to test communication with the sensors.
We have a small batch of prototypes running and I think that since v 1.04 the app reads data correctly.
Without the sensor the app just does nothing, however for now even that produces useful info, because we can fix a lot of minor glitches.

Post to thread

Message boards : Number crunching : nci not starting asap as expected


Main page · Your account · Message boards


Copyright © 2024 BOINC@Poland | Open Science for the future