We cannot see the page behind a link to anything on anyone's personal stats page. Thus, please post the distribution detail so we can assess the order of transmission/receipt.

The replication 4 is updated with each additional copy sent out, so by what you say it sounds like it was originally 1.

edit:

One in Six I've got presently in my queue shows a random verifier, up front, with just few minutes in between the issue:

Project Name: Help Fight Childhood Cancer
Created: 1/18/10
Name: HFCC_s1_02139897_s1_0001
Minimum Quorum: 2
Replication: 2

HFCC_ s1_ 02139897_ s1_ 0001_ 1-- - In Progress 1/20/10 04:10:58 1/30/10 04:10:58 0.00 0.0 / 0.0
HFCC_ s1_ 02139897_ s1_ 0001_ 0-- 610 Pending Validation 1/20/10 04:07:42 1/21/10 02:52:11 5.03 60.8 / 0.0

Looks like a tech has to explain this. It was an init 1. I speculate, one variation, that something happened with another result or results on the original device receiving the 0-- copy making it unreliable/suspect. That made the system decide to send out an extra copy which came back quickly. Then the rules realized a PV on it's own will never validate and send 2 minutes after 1-- two more copies, where one logically would have sufficed.

Possibly it's a device that detached, the odd thing being [as I understand it], that only when detached devices reattach, do the results get marked that way, else just going to No Reply after deadline. Thought until recently that detaching sent a signal to immediately cancel these results.

And, second speculation: The techs gave the remaining results of a slow batch an extra push, so that they quickest can send the results to the scientists? Not probably since it happened 2 minutes after the 1-- copy.

I'm sure there are more guesses to apply, so it's tech time to fill us in.

Which one is yours btw?

There were some work units that needed to be over scheduled because of how long they have been sitting on the grid. Some of these work units had been on the grid for 45 days and were causing entire batches to not be sent back. On Monday they were rescheduled and then given an extra push yesterday to help them finish quicker. There were probably about 100-200 of these work units corrected in this manner. I estimate that this these 100-200 work units were holding up over 200Gigs of results.

Hope this helps,
-Uplinger

The additional copies were sent to reliable hosts with only a 4 day deadline. So they usually return quickly.

-Uplinger