As usual Stanford has screwed up :P  take what I had posted on the first page and put the server that is not assigning units and place it's address IP into the command and you should not have any issues
 
If this is really working then on Windows the cmd looks like this:

netsh advfirewall firewall add rule name="FahClient workaround" dir=in interface=any action=block remoteip=171.67.108.xxx

To remove again:

netsh advfirewall firewall delete rule name="FahClient workaround"

So since these are the messages I am getting all night, guess what servers I am going to block for the next few days.

13:59:03:WARNING:WU00:FS01:Failed to get assignment from '171.67.108.45:80': Empty work server assignment

13:59:03:WU00:FS01:Connecting to 171.64.65.35:80
13:59:04:WARNING:WU00:FS01:Failed to get assignment from '171.64.65.35:80': Empty work server assignment
13:59:04:ERROR:WU00:FS01:Exception: Could not get an assignment
13:59:04:WU00:FS01:Connecting to 171.67.108.45:80
13:59:04:WARNING:WU00:FS01:Failed to get assignment from '171.67.108.45:80': Empty work server assignment
13:59:04:WU00:FS01:Connecting to 171.64.65.35:80
13:59:05:WARNING:WU00:FS01:Failed to get assignment from '171.64.65.35:80': Empty work server assignment
13:59:05:ERROR:WU00:FS01:Exception: Could not get an assignment

Thanks Ranmacanada for the reminder
 
 
What I'm wondering is why Stanford persists on allowing "Zombie Servers" (Empty work server assignment) to Prioritize themselves over Servers with Work that needs to be done ---> when work is requested by the Folding community ?
 
Just set the system up to automatically hide the "Zombie Servers" and STOP holding volunteer Folder systems hostage.
 
Stanford says that what we do is important ----> prove it Stanford, fix your system
 
Hey STANFORD ---> If you only Do what you did, You Get what you GOT !
 
OK, rant over. 
 
I'm tired and just trying to hit my personal milestone goal of 2 Billion in my first full year of Folding 
(I consider last Aug 19-31 my learning curve, and that I started Folding in earnest starting September 2016)

they will not get serious about it until REAL MONEY is on the line and they actually HIRE STAFF. they are solely reliant on grants and student volunteers it seems. until this changes they will not change as there is no reason for them to, other and these pesky mosquitoes buzzing around complaining about everything.
 
thanks for this update though, i had MANY idle slots and now i know why. though now i need to figure out an easy way to do this same thing through linux terminal.

The primary issue (and it's a LONG standing one) is that the assignment server and the work servers don't seem to communicate properly when the work server runs out of units - so even though the SERVER STATUS web site can figure out "work server has no units" the assignment server can't.
 
 Stanford has shown an amazing level of "I don't care" about issues like this (and the ALSO very long standing "client slot hangs if communication times out at the wrong time" issue), but they sure like to talk about "new features" when they bring out a new client version that doesn't FIX long-standing bugs.
 
 I also fail to understand why they have a command-line option (and use it BY DEFAULT) to that worthless "web interface" and no such option to the "advanced control" interface that actually IS usefull (though ALSO bugged, does NOT save slot information correctly when you have more than one slot in a machine, which I have ALSO reported to them quite a long time ago).
 
 I'm getting fed up enough with their far too frequent issues that I'm seriously considering telling FAH where they can shove themselves and move all of my current folding rigs to something else (BOINC perhaps, which also has issues but at least TRIES to fix them).
 
 Might be interesting to see HOW dominant I could get in DNet RC5-72 / MooWrapper if I pointed 24 NVidia 1070/1080/1080ti GPUs at it.