Problems with HDC units

[Me__2001]

i've started to notice recently that when i restart a HDC unit either after re starting my system or resuming them that they start from zero again ?

anyone else having problems?, i've seen feckits posts and i think he's having the same problem as me

 

[Anon_F]

I had to abort the unit i was crunching and changed the settings for which projects to run, but have one HDC sent before i reset them. You may need to do the same.

 

[Me__2001]

they complete ok, i just have to start again on them which is annoying as i'm wasting CPU time

 

[V_R]

Theres been a few probs like your ME, Do you just shut down the pc or do you close BOINC manually first? (Right click and close)

You should always close BOINC first give it a minute to write a checkpoint and then close the pc down.

Thats what i do and what WCG recommends and i have never lost any work. Try it and let us know.

 

[Me__2001]

i just shutdown and havent had any problems in the past, it does explain why after re starting but not after suspending boinc and then resuming them ?

 

[V_R]

I never use suspend myself, so i dont know about that but there is a thread over at WCG about this have a look here.

The most recent post says they are aware of a problem with a fix due out soon.

 

[Me__2001]

thanks for the link, guess i'll just have to stick with it for now

 

[V_R]

We now have an offical update regarding this.

Thanks to the help of our members we have identified a problem with checkpointing in the Help Defeat Cancer application (on both BOINC and UD). We are currently testing the fix for this in development. Once this looks good we will begin beta testing and if that goes well release it into production.

Until the fix is deployed we have disabled the code that would only grant half credit to users who claimed high. This is only disabled in the Help Defeat Cancer application.

Thread here

 

[Me__2001]

thanks for the update V_R

 

[Anon_F]

Aaaaaarrrrrrgggggggghhhhhhhhhhhhhhhhhhhh

Spoke to soon HDC unit crashed after 3.3 hours & have had to abort it
Now back to FAAH units

 

[itsme]

Whoops running HDC! my temp rises from 39c to 51c Me thinks no more HDC utill it`s sorted back to Faaaaaaaaa............

 

[V_R]

Hey feckit, by what do you mean crashed? BOINC does do some funny things sometimes, I wouldn't be in too much of a rush to just abort a wu, unless your certain its gone tits up.

 

[V_R]

Whoops running HDC! my temp rises from 39c to 51c Me thinks no more HDC utill it`s sorted back to Faaaaaaaaa............

Hows that going to make a difference?? Its still gonna use the same cpu still at 100%. You sure you were running it before and not at idle?

Or just turn the heating down!

 

[Anon_F]

It did sadly go "tit's up"

Hey feckit, by what do you mean crashed? BOINC does do some funny things sometimes, I wouldn't be in too much of a rush to just abort a wu, unless your certain its gone tits up.

It was using 255MB of HDD, the % crunched was going up & down like a yoyo & time to completion was also going up, not down!
I kept checking and after 1 hour decided to abort, hav lost 10 hours of crunching from two crapped out HDC units.
But now back to FAAH Units

 

[V_R]

Anything in the BOINC log?

 

[Anon_F]

Lots of bits of red writing, once aborted everything now back to normal

 

[V_R]

Lots of bits of red writing, once aborted everything now back to normal.

Good, thats what i was hoping you'd say, Remember the 'CPU time' and 'To completion' figures are only estimates.

 

[itsme]

Hows that going to make a difference?? Its still gonna use the same cpu still at 100%. You sure you were running it before and not at idle?

Or just turn the heating down!

Don`t know why but Faah Units run at 39c no problem HDC run at 51c using the same CPU 99%>100%.???????????? when this one unit has finished I will let you know what happens
Heating..... whats that