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

Quick Go ยป
No member browsing this thread
Thread Status: Active
Total posts in this thread: 2370
Posts: 2370   Pages: 237   [ Previous Page | 190 191 192 193 194 195 196 197 198 199 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 1709958 times and has 2369 replies Next Thread
mikey
Veteran Cruncher
Joined: May 10, 2009
Post Count: 824
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: New thread for It's raining Dengue, Hallelujah!...

I just changed several machines over to Dengue Phase 2 and 1 machine got lots of work, I set my cache to 1.5 days, but the 2nd machine got the message 'no work is available for Dengue'! It can't be over ALREADY can it?!
----------------------------------------


[Jan 17, 2012 2:04:27 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: New thread for It's raining Dengue, Hallelujah!...

Aaron,

Any type of illogical task hopping is logged in the message/event log, to include memory and disk / swap file shortage and (virtual) memory shortage is sure one other reason for weird behavior, on the surface.

My machines I've just set to allow swap file growth ad lib, and 75% of that granted to be used by BOINC. Swap files are though set to not shrink below 1.5 times RAM. As with Linux where swap files are placed in their own partition, on Windows I've set same to minimize chances of fragmentation. With oodles of space on modern drives, no issue to reserve 25Gig, so the chance with an octo of running into the absolute wall is minute. BOINC data-dir also has it's own logical drive... easy to manage from a permission / security exemption perspective and working also to reduce chance of drive fragmentation.

Unplugging to let redundant tasks still compute I'd not advocate. Chances are that a task has already been migrated to the master DB [it's quick these days] i.e. is no longer on the Result Status page by the time you come round to plugging in again, you could end up getting a ''too late / no longer on DB" meesage. Then at least use the client scheduled network preference page, to reconnect when you know all have started. The server does not instruct aborting for running tasks, unless known to be bad.

For now, on DDDT2, got 74 in PV jail with a mean return time of 25.6 hours. If at all the wingmen could get me a different form of precipitation... of validated tasks.

--//--
[Jan 17, 2012 2:06:52 PM]   Link   Report threatening or abusive post: please login first  Go to top 
bieberj
Senior Cruncher
United States
Joined: Dec 2, 2004
Post Count: 406
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: New thread for It's raining Dengue, Hallelujah!...

I just changed several machines over to Dengue Phase 2 and 1 machine got lots of work, I set my cache to 1.5 days, but the 2nd machine got the message 'no work is available for Dengue'! It can't be over ALREADY can it?!


I am still getting work so it is not yet over - from what I see (at b093 as of 90 mins ago) - about 60% of the work has been downloaded so there is still 40% to go.

Either be patient or keep hitting the update button every 20 seconds until you get some work.
[Jan 17, 2012 2:26:38 PM]   Link   Report threatening or abusive post: please login first  Go to top 
TXR13
Cruncher
Canada
Joined: Dec 5, 2005
Post Count: 36
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: New thread for It's raining Dengue, Hallelujah!...

@Ascholten & SekeRob

I also experienced the beginnings of a BOINC panic on one of my DDDT2 crunchers. Checking the cache size, I noticed that it was now awfully close to deadline limit. Why, I couldn't tell you--from my calculations, the machine should be clear that amount of work in half the time at the most.

To give it some breathing time, I suspended work fetch and suspended half the WUs. Overnight, it seems the duration correction factor fixed itself, and the machine now correctly realizes that it can complete everything in well under deadline, and FIFO processing has resumed. I've re-enabled work fetch and resumed the suspended WUs, and everything is ticking along nicely again.

Still, something to keep my eye on...
----------------------------------------

[Jan 17, 2012 4:48:04 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: New thread for It's raining Dengue, Hallelujah!...

I checked my results / pending results and have seen a significant amount of tasks that ended up in 'error' status. Is this normal or is something going on on my end I wonder.
I just hate to do all this work and have it be worthless to the guys on the other end.

Aaron
[Jan 17, 2012 10:57:30 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: New thread for It's raining Dengue, Hallelujah!...

Aside for the occasional job that bombs at / near first checkpoint on all copies like the below log shows [had one in hundreds], there is maybe the occasional download fail driven error. If your wingmen process normally, it's your system having the issue.

--//--

<core_client_version>7.0.7</core_client_version>
<![CDATA[
<message>
The system cannot write to the specified device. (0x1d) - exit code 29 (0x1d)
</message>
<stderr_txt>
INFO: No state to restore. Start from the beginning.
CHARGE OUTSIDE INNER GSBP REGION
Encountered error. Exiting.

</stderr_txt>
]]>
[Jan 17, 2012 11:31:04 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: New thread for It's raining Dengue, Hallelujah!...

How far we are in distribution:

7192 World Community Grid 18-1-2012 9:56:20 Finished download of dg05_b_g23_b221_to_b229_typeC_solv_repu_0.8_0.9_b1.prt.gzb

b481-b490 is the last batch, and b491-b500 if Seippel got the load fixed of that batch. See http://www.worldcommunitygrid.org/forums/wcg/printpost_post,359702

So far, 6 days into this run 71,753 results have validated. Not 25% of total.

Come on guys, this is what we do [best]... Crunching. biggrin

--//--
[Jan 18, 2012 9:08:21 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: New thread for It's raining Dengue, Hallelujah!...

Thanks for the info biggrin
[Jan 18, 2012 9:30:35 AM]   Link   Report threatening or abusive post: please login first  Go to top 
dango
Senior Cruncher
Joined: Jul 27, 2009
Post Count: 307
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: New thread for It's raining Dengue, Hallelujah!...

It looks, that everyone fetch the much as he can :)
within few hours the raining stops, but results will come next 7 days :)
[Jan 18, 2012 9:46:50 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: New thread for It's raining Dengue, Hallelujah!... [2nd Edition]

Correction [now on b_g25]: Think there's a good day left as the amount of tasks put in the feeder are restricted [hence why the many ''no work'' between successful batches] fetches, to give allot more volunteers chance of grabbing some, even those not at the button with custom profiles set.

--//--
----------------------------------------
[Edit 1 times, last edit by Former Member at Jan 18, 2012 12:07:47 PM]
[Jan 18, 2012 10:12:36 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 2370   Pages: 237   [ Previous Page | 190 191 192 193 194 195 196 197 198 199 | Next Page ]
[ Jump to Last Post ]
Post new Thread