Server Problems? - currently the U/L one

Problem fixed 23:07 UK time, and the Feeder is now running.

After some discussion, we've decided to move the credit generated per workunit back to where it was, however reduce the maximum possible credit generated per hour in half -- this should keep our project in line with other BOINC projects, but not penalize people not running optimized apps. These changes should effect all newly generated work so it might take ahwile for everything to propagate out.
--Travis

---

Don't drink water, that's the stuff that rusts pipes

Uploads are now fine -- no work available though.

---

Uploads and reporting results are fine for my rigs still with sork in their short caches.

Server status shows everything running OK, but no new work avaliable. Also the work in progress has fallen from a peak of ~72K to the current 49.7K.

I wonder if this is preceeding Dave's indication the servers were planned to go down for some tweaking, prior to the release (later this week) of the new MW faster stock client?

new work is coming through. i don't think they're in a hurry to release the new app, since the unofficial optimized one is out: they better take this opportunity of a 'hybrid' population to anticipate the server load/bandwidth when everybody will use a fast app. these last days showed it's not a completely smooth transition, WUs being computed faster than they are generated!
let them balance things out, then focus on basically making the fast app official. it's not a race, there's no deadline.

Anyone having problems with their WU results ready to report?

I am having to manually force mine to get to acknowledged.

Looks to me like the server is struggling to keep up. Some of my machines are having problems getting new work since the server is not keeping up with the demand.

---

The interesting thing about WUs waiting to send, when they are available to download, is there in no point in extending the cache. There is a 20 WU per core/CPU cache limit which the system will not stray beyond.

Best have a back up project running concurrently

It looks like there may be a problem with the upload server?

It seems reluctant to accept results during the upload cycle ATM. They are repeatidly asking to report the efforts of crunching.