Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 49
Posts: 49   Pages: 5   [ Previous Page | 1 2 3 4 5 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 54046 times and has 48 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

Hi armstrdj - I have been running on 4 PCs with 100% CPU for the past hour with no stalling. I am going to leave these 4 systems (2 are 64 bit Win 7, and 2 are 32 bit Win 7) running overnight and see what the morning brings.
Till then ...........
----------------------------------------
[Edit 1 times, last edit by Former Member at Sep 20, 2011 8:37:42 PM]
[Sep 20, 2011 8:31:06 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

Question I have with this throttling i.e. have a CPU time% setting below 100% is, if the "Leave Application in Memory when preempted" is on or off. Normally it would be irrelevant, unless there is an underlying bug in BOINC itself, not retaining the app in memory during the cool-down phase... at 50% 1 second off, one second on.

Also, is the "While processor usage is less than xx percent (0 means no restriction)" set to zero or some other value?

Personally, if the objective is to have more power to the user rather than running the machine cooler (the objective of the CPU%), I'd only be running the "While processor..." option and set CPU time % to 100.

--//--
[Sep 20, 2011 9:43:53 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

Good morning folks - well, I have now been running with 'while processor usage is less than 0 percent' and 'use at most 100% CPU time' - no stalls on any of the PCs running at 100% (however, also no stalls on the one remaining PC running at 70% - so not conclusive I'm afraid) - I have usually had 3-4 stalls per day across the 5 PCs. I will let the 100% CPU continue for now. (leave applications in memory is un-ticked in all PCs).
[Sep 21, 2011 6:15:36 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

Raul, it's better to set the "while..." to zero if it is at 100% to disable the function, as even at that ''logical, give it all'' value BOINC tries to pause the science at times [so observed in past testing] when the system is really busy, then with LAIM off the sciences could unload, IIRC]. BOINC sciences will slow down anyhow to only use all spare cycles. Generally if your systems are 24/7 I'd recommend LAIM on as rush jobs that might come could suspend these tasks, and then revert to last checkpoint when resuming. On DSFL I've seen 1 hour and more at times for the heavy job steps.

Wonder if the chance of WUs getting stuck increases with the length of these tasks [the number of jobs included], or it being pure random to happen, think latter from reports.

--//--
[Sep 21, 2011 8:01:59 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

Hi Sekerob - the 'while....' is set to zero, and the 'use at most....' is set to 100.. My systems are 24/7, so I'll set LAIM on as you recommend.
Thanks for your help - much appreciated
[Sep 21, 2011 9:06:43 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

Well, I've been running for almost 24 hours at 100% CPU with not a single stalled task - seems too good to be true. My one PC that is still running at 70% has also not had any stalls, but it's a slower machine and has only run 2 tasks in the past 24 hours. I would say that the problem has been identified - what would you like me to try next?
----------------------------------------
[Edit 1 times, last edit by Former Member at Sep 21, 2011 6:01:33 PM]
[Sep 21, 2011 6:00:26 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

I have a stuck wu:

DSFL_ 00000020_ 0000038_ 0174_ 0--

It has been running for 9 hrs (elapsed time), and it is at 58.750% not moving as progress but (it is now suspended) if I restart the elapsed continue to tick, but no progress.

By the way, I cannot see the properties of any wu on this machine, I do not know why, every time I click on properties (Win 7) it seems the little window appears but I am not able to see it (it is just shown as miniature on the task bar). Any suggestions?

What should I do with this wu?

Thanks
[Sep 27, 2011 9:03:18 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

Well, after running for 8 days without any problems, I reset Boinc preferences to not use more than 90% CPU - guess what, I got a stuck task within an hour. I reset preferences to use 100 CPU, rebooted, and viola - all running well again. I shall continue to run at 100% for as long as possible, unless anyone wants me to try something different.
latakia - I suggest you set Boinc preferences to 'use no more than 100% of the processor', then close everything down and reboot - the task should run again and complete normally, and I dont think you'll get any more stuck tasks as long as you keep running at 100% CPU - watch your PC's temperature.
----------------------------------------
[Edit 2 times, last edit by Former Member at Sep 29, 2011 4:19:57 AM]
[Sep 28, 2011 9:58:15 PM]   Link   Report threatening or abusive post: please login first  Go to top 
armstrdj
Former World Community Grid Tech
Joined: Oct 21, 2004
Post Count: 695
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Stuck workunits

Raul, which version of the BOINC client are you running?

Thanks,
armstrdj
[Sep 29, 2011 1:32:41 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Stuck workunits

6.12.33 (both 32 and 64 bit)
----------------------------------------
[Edit 2 times, last edit by Former Member at Sep 29, 2011 5:06:31 PM]
[Sep 29, 2011 5:04:47 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 49   Pages: 5   [ Previous Page | 1 2 3 4 5 | Next Page ]
[ Jump to Last Post ]
Post new Thread