[Edit]ok, different request, for a pointer to the most succinct description of the quota algorithm. From what I'm seeing I'm guessing an accidental abort of some tasks from another project (or rather, a deliberate abort after accidentally selecting a project I didn't want to crunch) has made this host non grata for all projects none of which will give it work. It's going to run out of work before too long, is it going to be forgiven in 24 hours?

It will be forgiven. I can't remember if it happens at 12:00 GMT or 00:00 GMT, but it does happen.
At that point it will get new WUs and start building back up its reliability rating.

hmmm, succinct, hmmm, maybe if you translate the verbose from this FAQ: http://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,28636

If you have work still to crunch through and it turns valid AND you have not used up the 80 WU max per device core per day for a previously reliable client then you could get more today. If not, then there has to have been a bunch of 'accidentals' going on. If then, send a deep grovel mail to the support contact [to include a full technical explanation] and maybe they'll reset the quota for you... Else, the wait is on till 00:00:01 UTC to get 1 to proof worthy. 80 is allot which is 320 for a quad, my 64 bit linux managing to do about 65-70 HCC jobs in a day i.e. never ever getting close to the core quota.

Someone who wishes to build a 10 day HCC cache will need initially several days to get there. With a sample of 70 *10 / (4*80) = 2.2 day, but since 65-70 will have been completed on the first and second day, it's only on the 3rd day that full stock can be had. 10 days is though not recommended for various issues, mostly because just the smallest downtime will cause the last tasks to go overdue and get auto-cancelled by the newer clients which goes against the daily quota.

That makes the most sense, just commenting on what I read elsewhere.
Thanks for the pointer to your earlier post, it's clear enough. Of course I made my mistake right after UTC midnight, but finally have returned enough work to get a small trickle, enough to get the host through the night. Certainly want to have some quota left when more DDDT-2 work arrives!

Though this host probably won't go back into the reliable category any time soon. I'm sure others have also wished for the relevant host stats to be visible on the site and this isn't the place for the topic, but thanks for answering my question anyway.

Thanks Much for the info. I was trying to manage my queues to get some more DDDT2s and fubared it. So now 3 or 4 of my super reliable machines will not be so rated for a bit. Mayhaps that is why I did not get any DDDT2s last evening even though my trap was set(for the first time).