Pool node priority is not working (active/passive config)

Hi guys. I'm trying to work through an issue. Currently, I have a pair of standalone MS ISA 2006 Servers that I want to run in an active/passive model. They are the front-end authentication mechanisim for our SharePoint Farm (5 Web Applications contained within). I have 5 IP addresses on each ISA Server, each active one paired up with it's passive counterpart in their own respective pools/Virtual Servers. I need to come up with a way to force all of the traffic to go to the active node and only go to the passive node if the Active node fails.

The problem that I'm having is that 2 out of the 5 apps seem to be dumping the requests to the passive node, no matter what I do.

This is the layout that I have for one of the problem web apps (I have triple checked via the web UI that they are all identical).

   
 VIRTUAL ADDRESS xxx.xxx.xxx.125   UNIT 1   
 |     ARP enable   
 |     (cur, max, limit, tot) = (0, 21, 0, 259)   
 |     (pkts,bits) in = (4543, 6.687M), out = (5862, 50.10M)   
 +-> VIRTUAL VS_ISA_APP1_COMPANY_COM_HTTP   SERVICE any   
 |     PVA acceleration none   
 |     (cur, max, limit, tot) = (0, 21, 0, 259)   
 |     (pkts,bits) in = (4543, 6.687M), out = (5862, 50.10M)   
 |     requests (total) = 0   
 +-> POOL NET_ISA_APP1_COMPANY_COM  LB METHOD round robin   MIN/CUR ACTIVE MEMBERS 0/2   
 |     (cur, max, limit, tot) = (0, 10, 0, 68)   
 |     (pkts,bits) in = (1601, 2.497M), out = (2229, 18.68M)   
 +-> POOL MEMBER NET_ISA_APP1_COMPANY_COM/xxx.xxx.xxx.28:any   active,up   
 |   |     session enabled    priority 100    ratio 1   
 |   |     (cur, max, limit, tot) = (0, 6, 0, 34)   
 |   |     (pkts,bits) in = (1319, 2.038M), out = (1982, 17.90M)   
 |   |     requests (total) = 0   
 +-> POOL MEMBER NET_ISA_APP1_COMPANY_COM/xxx.xxx.xxx.29:any   active,up   
 |     session enabled    priority 10    ratio 1   
 |     (cur, max, limit, tot) = (0, 4, 0, 34)   
 |     (pkts,bits) in = (282, 459424), out = (247, 786048)   
 |     requests (total) = 0   
 

Configuration steps I've done to try and make this happen:

* I've set the priority on the active node to be 100 and the passive node to be 10. Load Balancing method is round robin.

* Following this guide, I set up a single node persistence profile and applied it to each of the Virtual Servers. http://devcentral.f5.com/Wiki/default.aspx/iRules/SingleNodePersistence.html

* I tried messing around with Priority Group Activation, but it didn't seem to affect anything. I tried it with 3,2,1,0 and it nothing changed. To be honest though, I'm a little fuzzy on how that whole thing works.

Now like I said, this is working perfectly and consistently for 3 out of 5 apps. Up until yesterday afternoon, it was actually working for 4 of them, but an additional one started flipping, which led me to believe that there is a flaw in my configuration and I'm getting lucky on the ones that are working correctly.

This is a 9.4.3 LTM, by the way.