GNU/Linux S5R3 "power users" App 4.35 available

Wedge009
Wedge009
Joined: 5 Mar 05
Posts: 128
Credit: 17551695717
RAC: 6974308

Regarding the 'stuck'

Regarding the 'stuck' work-units I mentioned earlier, I can confirm that it occurs on two of my hosts now. The common factors here are the Linux 4.35 version of the Einstein application, and BOINC 5.10.8. It may very well be a BOINC-specific issue, but I have not seen this problem occur for any other project or on my Windows hosts.

Soli Deo Gloria

Gary Roberts
Gary Roberts
Moderator
Joined: 9 Feb 05
Posts: 5874
Credit: 118176910210
RAC: 23875668

RE: Regarding the 'stuck'

Message 79558 in response to message 79557

Quote:
Regarding the 'stuck' work-units I mentioned earlier, I can confirm that it occurs on two of my hosts now. The common factors here are the Linux 4.35 version of the Einstein application, and BOINC 5.10.8. It may very well be a BOINC-specific issue, but I have not seen this problem occur for any other project or on my Windows hosts.

I've been seeing stuck tasks just like you describe for over a year now. I posted about it just very recently here. In my case the behaviour is on both Windows and Linux and on a variety of different BOINC clients and science apps. Flakey hardware and heat seem to be the two common ingredients. I don't notice the problem on very good hardware - servers for example - no matter how hot they get.

Cheers,
Gary.

Wedge009
Wedge009
Joined: 5 Mar 05
Posts: 128
Credit: 17551695717
RAC: 6974308

That's interesting. However,

Message 79559 in response to message 79558

That's interesting. However, I have only noticed the stuck work-units with this version, plus I am quite certain my hardware is stable. None of my hosts are overclocked, and they have been working reliably for several months. Maybe the similar symptoms we see are due to different reasons, or maybe not. I only posted what I observed here in case others have noticed a similar problem.

Edit: Also, heat does not seem to be an issue - we're just coming off an unusually cool summer, and even with this last gasp of warm weather, ambient temperature was no more than 25C when the work-units got stuck. (CPU temperature is stable at around 50C which I believe is quite cool for the old Barton-core Athlon XP.)

Soli Deo Gloria

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.