I have a quick question about how netmap will behave in a certain situation. Lets say I have a 1.1.1.0/27 that I want to netmap so that it appears that certain clients are coming off of different public IP addresses. So i set up a rule that will netmap the subnet 172.16.0.0/24 to 1.1.1.16/28 (I don’t care that they are of different size) of the IPs that I have. How will the Mikrotik handle the netmap rule for the broadcast IP of the /27? Will it skip over that IP since it is in use by the broadcast, or will it sometimes try to nat a person out of that IP, and make it so that they cannot get online?
Anyone know?
http://wiki.mikrotik.com/wiki/Manual:IP/Firewall/NAT#1:1_mapping
You specify a range for netmap, not a network address with a netmask. Just specify the range so the broadcast IP address isn’t included. Also, netmap maps 1:1. I don’t know for sure what it’ll do if one range is larger than the other, but it probably wouldn’t be doing what you want it to do.
Netmap won’t take anything other than the full subnet (broadcast and network IP included) in the version we’ve been using 4.5. I’ve tried specifying the range without them, and it refuses to take it both through Winbox and the command line, so i’m guessing they changed it. From what I can see it works fine having one subnet larger then the other, it just loops the IPs back, which is fine by me. It also seems to me to work a lot better than specifying a range with action=scr-nat for allowing traffic to pass and work.
Does anyone know?
How about this, if i set the action to scr-nat and put in a range of IP addresses, will an individual client always go out of the same IP within that range, or will different connections choose a random IP to go out of?
I believe that if you need the same address you need action=same
I would actually just like to make sure that with that action and rule, a guest with an IP of 192.168.50.25 will always have their connections NATed out of x.x.x.125, and it won’t choose a random IP for each session.