I returned an error NATAN seven hours ago. Nothing since. Tried several "Update"s. GPU getting a well-earned rest!

Is it me or thee??

---

'twas me! Call off the dogs! After a reboot I got a TONI. Strange...

---

Is feeder about to run dry ??
2/9/13 11.45am (+8 timezone)
tasks ready to send gone done to only 2
5 minutes ago was just 4.
Yesterday was huge for the grid I know...
2 users with over 4,000,000 didn't even make top 5.
Top 10 users were all over 3,000,000
BUT has that big day put the tasks available into "unavailable"
Maybe I'm reading it wrong ??

Had a power outage today for 9 hours, the forest fire burned the power lines so it wasn't me.

I think I'm nearly at critical point for no work on my number 1 host which runs 6 tasks at a time.
Currently has...
3 tasks uploading
1 task at 98% and another at 78%
4 tasks working along under 40%
Only 1 task in cache !!
At least they are all long runs but soon nothing up it's sleeve.
and server status shows only 1 task available to send.
I'm guessing can't download until uploads finish ??

Normally a computer with 6 GPU's can download up to 12 GPUGrid WU's - 6 running and 6 waiting to run. Depends on your work buffer though. If it's set low (0.01 days) then you will probably only have 6 or 7 GPUGrid WU's at any one time. If it's set to a day then you should have close to 12 WU's most of the time. Obviously if there is a WU shortage server-side you won't get 12 WU's. Also, if you get lots of failures you won't get many either.
At present there are issues with the estimated runtime of some WU's, which might be over-estimating runtime by a long way. This may also be preventing you from getting a full complement of WU's.

---

FAQ's

HOW TO:
- Opt out of Beta Tests
- Ask for Help

Cheers,
I.m pretty sure that atm is WU shortage.
I usually have 6 + 6 WU's with buffer at 1 + 1 days additional.
Completion time is always good - 3 x 690 cards.
and error rate since cuda5.5 and new drivers has been pretty much nil (maybe even 0)

Just checked and if I read correctly (lower box) o tasks available...(for non-Betas)

application unsent
Short runs (2-3 hours on fastest card) 0
ACEMD beta version 1
Long runs (8-12 hours on fastest card) 0

Or is it this that is corret in upper box on server status page...

Tasks ready to send 1

Can someone please explain how the scheduling priorty is calculated and what the effect on tasks being sent is. mine is -2.0 on the host.
Also my duration correction factor is 1.1655 on that host.

My task time estimation seems acceptable prior to commencing - long Nathan says 11:04.20