Forum Discussion
Nimbostratusreverse proxy
NacreousOk well it seems you don't quite understand the guide. Let me summarise for you.
Rewrite profile = URI Translation
HTML profile = Tag Removal (optional)
Local Traffic Policy (LTP) = pool selection
The URI translation modifies the data however it does not choose a destination for you. The load balancing method on your pool does that. But in this case we want to override that with LTP. You can instruct LTP to select an individual pool member but multiple pools is a better approach. Separate your servers into pools, use LTP to select the pool you need based on the URI.
So why does it work this way? The rewrite engine is designed to reformat data and that's it. LTP on the other hand is designed to to a whole lot of things including choosing a destination based on that data. Pools are better because if you decide to add another server to support capacity then you just add one more to the relevant pool. Also persistence methods such as cookie persistence are often linked to the pool. In the case you want persistence to work across pools as you might here, their is a tick box for it.
NimbostratusI split out the pool into 3 seperate pools. I created single policy with 3 different rules (one for each pool). I modified my HTTP VS to use no default pool and to use my policy (is that correct)? In addition I assigned a client SSL profile to the HTTP VS (required to assign my policy). Do I need an HTTPS VS since the nodes are using ssl?
in the LTP rules should the HTTP-host be what the host the client will see or the address of the pool member (I think it would be what the client will see not the node).
