Very early this morning I changed my general preferences to up the caches from 0.1 days to 1 day.
From memory this was done on the seti site.

This seemed to corrupt all my clients as many hours later I discovered that none of my farm was even requesting work from any project.

I was able to confirm that machines could contact both predictor and Einstein (seti was down at the time anyway) as the project percentages were updated correctly (some machines had only just been attached to Einstein) and the credit totals for both projects updated correctly. But boinc would not even request work , despite the machines being completly out of work.

I tripple checked that the preferences were set to ask for work etc .

I then used Einstein to change the cache preferences to be 2 days instead of 1 day . I then manually refreshed each machine to contact either Einstein or predictor (the new preferences propogated over to predictor ok) and now each machine was happy to request work from all projects.

I am guessing that due to the issues seti was having , it managed to somehow corrupt the preference details ?

Due to the number of machines in my farm it was a right pain in the rear to correct this , but all is well now.

So if anyone else has the same problem , you can try the same to see if it clears it.

please , please, pretty please can we have the option to set project percentage shares and cache settings localy for each machine. i.e a setting that says ignore any global settings and overide them with these local settings as 3 venues is somewhat limiting when you have a lot of different machines.

Shady

---

<img src='http://www.boincsynergy.com/images/stats/comb-1527.jpg'>