Here's [TA24506|http://support.apple.com/kb/TA24506], with an introduction to Xserve LOM hardware.
Yes, you can have both ports active for Mac OS X Server data, and you can also have both RJ45 ports active for LOM access. And yes, all in parallel. If you have both ports active for data on the same LAN, you will likely want to have the IP addresses in different subnets on the LAN, or IP routing can potentially get confused. Same for the LOM configuration; active on two ports is best in two subnets. Right now, setting up IP subnet routing might be a bit more advanced than you want to get into, so the simplest answer here is probably "don't try that (yet)".
For now, it might be simplest to have one gigabit port for data, and the other for LOM, and two cables. Or you can have one data and one LOM IP on the same connection, on one cable. The former can be a management LAN, the latter can be the typical shared LAN shared-switch network configuration.
Ok, as for your "but cant fugure out where too get the ip address", this could be interpreted as generic IP confusion (eg: what's a static IP address?), generic Ethernet confusion, confusion over the ability to have multiple IP addresses on one Ethernet connection (RJ45 or otherwise), or that there are effectively two different network devices present within each Xserve RJ45 physical connection.
Or this could be confusion over how you address the LOM locally, and the answer to that is you use the local graphics display or screen sharing to get to the Xserve box itself, and then you
always specify accessing the LOM via the so-called local path when you're initially configuring it.
As for static IP, LOM requires one (or two) IP addresses (for activating the LOM on one or both of the RJ45 connections, and one or more IP addresses for the Xserve host activities on the RJ45 physical connections.
And what static IP addresses to use is up to your local network administrator. If that's you, then you will want to assign those static IP addresses outside the pools of IP addresses used by whatever DHCP server(s) or VPN servers might be on your LAN. I usually set the static IP addresses to the lower ranges of IP addresses and the dynamic IP addresses (the "pools") to the higher address ranges, as that (largely arbitrary assignment) saves (a little) on the typing, and as the smaller numbers tend to be easier (for me) to remember when I'm setting up stuff. Outside of the necessity of avoiding address collisions and (typically) configuring all your boxes within the same IP subnet, how you address your local network stuff is entirely up to you.
