Greetings!!
Thanks in advance for your time and explanations..

This is a puzzler.
I'm running BOINC 6.6.36 and some WU get aborted before the deadline.

I have signed up for 3 projects - but i usually only run one at a time.

I am running currently running WCG with 7 WU - each with around a 5 hour estimated time to complete.

The scheduled report deadline is 8/20.
It is now near noon on 8/17.
Of the 7 WU, two are active with
a) 44 minutes estimated completion
b) 2 hours and 21 minutes to complete.

The third WU has a status "Aborted by project"

The 4th, 5th, 6th, and 7th WU are ready to start.

I'm running on a laptop with a dual intel cpu.

I figure that the work should be done in
10 to 12 hours.


I travel twice a week and have the laptop off for 3 to 4 hours while I'm in transit. I had it off for about 3 hours last night.

I'm not sure when the WU was aborted - sometime between noon yesterday and noon today.

--
On a different PC, I was running SETI@Home and had the same problem. I did an 'update'. The aborted WU were reported and more WU were downloaded!!

A complication on this one: I had set the local preference for network usage: connect every ____ days
to 5.0 to build up some work for the week; then reset the connect every ____ days back to 0.75 (my default)
I only get to this PC twice a week and do not leave it connected.


Question:
What causes the abort and how can I avoid sending back aborted WUs???
I suspect that I'm misusing the connect time setting - somehow.

Thanks again,
Jay

There are basically two abort by project tools available to the project administrator.

1 - Abort WU that hasn't started to be processed yet. This can be implemented by a project if a WU has already been validated by others that received the same work. This is so you don't waste CPU time on work that is redundant and is no longer needed. This is most likely to be seen on system with a large cache of work.

2 - Abort WU even if it is already being processed. This is used extremely rarely. It's a tool available if a project discovers a major flaw in a batch of WU or in the application software making the work obsolete. It could also be used if a project is at the end of it's life and additional work would be of no use.

---

Boinc V 7.4.36
Win7 i5 3.33G 4GB NVidia 470

Thank you Aurora B. and Sekerob....

Ah-Hah!! You are right. Thanks for the explanation.
I should have looked at the WCG status first!


https://secure.worldcommunitygrid.org/ms/device/viewWorkunitStatus.do?workunitId=88616012

It is as you said.. Very interesting...

Result Name: CMD2_ 0027-2RB4_ A.clustersOccur-3CJC_ D.clustersOccur_ 0_ 3056_ 6171_ 2-- 
App Version Number: 614  	

Status          Sent Time         Time Due/Return Time     CPU Time (hours) 	
------------------------------------------------------
Server Aborted 8/16/09 04:02:34   8/17/09 16:28:43 	   0.00 	
----Valid----- 8/06/09 03:56:24   8/17/09 05:02:23         3.42
----Valid----- 8/06/09 03:51:19   8/07/09 08:55:37         5.05

I'm amazed at how well it works.

Thanks again,
Jay