No work units from Einstein

I also started not getting EAH wu a couple days ago..it seems as though the problem began as soon as I started crunching with Albert. It shouldn't be a debt problem..i've been flipping smooth as silk between SAH and EAH on an hourly basis since LAH ran out of work. I got the utility..does this indicate a debt problem? If so, I can't understand how i built one up..
PRJ: EINSTEIN@H STD: -1043.563368 LTD: -586311.704861 RSRC: 100 -------------------------------
PRJ: LHC@HOME STD: 0.000000 LTD: 306468.516308 RSRC: 100 -------------------------------
PRJ: SETI@HOME STD: 1043.563368 LTD: 279843.188553 RSRC: 100 -------------------------------

One other odd thing..on my comp host page the 'time work allowed' is showing as .0015%..this is a big change and yes I DO have 'run always' set. I HAVE been crunching the SAH wu in my cache at a normal rate and in fact have quite a few queued for reporting and upload from the last days outage. I've been advised to do the following: Easiest fix... shut down BOINC completely, make a backup, then open client_state.xml with Notepad, about 30 lines down is "active_frac" - it'll say 0.0015, simply change that to a 1, save, and relaunch BOINC. It'll drop a bit from the 1 over the next few hours, but if you truly are running 99% of the time, it should only drop to .99.
Update..my boinc just switched over and started crunching on my last Albert WU for the first time in several hours...strange things are happening..

This is absolutely, totally, unacceptable. If it _is_ time to remove the app_info.xml file, and therefore the "beta" Einstein apps, then the project needs to say so.

There is nothing on the home page. Nothing in the thread on the "Albert" application that is referenced from the home page. And, just to make things _more_ fun, there is a possibly show-stopping bug in the Linux version of the Albert application.

The use of the beta application was routinely recommended to people experiencing the graphics bug of the old einstein application. Many of those people do *not* know what it means to run a beta software. Thus I'd consider Bill's complaint as a valid one.

There seems to be an issue with the Linux version writing to the hard disk too often in some cases. See threads albert writes to disk too often and New (Albert) application and workunits

If it was an optimized application obtained from a third-party, as at SETI, you would be absolutely correct. But in this case it is a "fix" provided BY THE PROJECT, and heavily recommended to just about everyone. I know that the app_info.xml file has been discussed (how to either automatically delete it or tell people that they NEED to delete it) by Bruce Allen on the BOINC developer mailing list. They knew a week ago that WHEN they stopped issuing Einstein WUs, they needed to "do something" about the xml file. My problem, as someone who spends a lot of time on many boards, attempting to offer help and support, is that a participant asked a question, and RELYING ON THE INFORMATION I HAD, I gave them the "wrong" answer.

If I, who spend several hours a day on the boards and follow all of this very closely, had no idea that the project had stopped issuing new Einstein work and was only issuing Albert work, how is someone who started BOINC three months ago and hasn't looked at it since supposed to magically know they need to do something?

Bruce, if you read this - my vote is to send the delete signal. It may not be "expected" by the rules, but this is a special case and it's the project's responsibility to take care of it.