I picked up one of these at NCIX on Saturday... - Installed it Saturday evening, getting it working as a backup wireless router; - Upgraded the firmware from TP-Link, which required redoing that configuration effort :-( - Pulled the OpenWRT binaries (per Giles' link), and installed, which was, modulo a wee bit of "wait afterwards hoping it worked", about as easy as he suggested - Introduced configuration to have it take over my WAN link, which worked, but then didn't seem friendly to needs for incoming connections (torrents^H^H^H^H^H^H^H^H and ssh), so plugged the old router back into place I poked at sundry sources of docs concerning the differences between firewall rules and firewall routes, which didn't really tell me anything I didn't already know or hadn't already done. Yesterday, I wrote up a script to pull config tarballs off the router and throw it, version-controlled, into a favorite Git repo, so I have backups. In keeping with the strategy I described a bit at the last meeting; anyone interested can review the GTALUG "backups" repository at GitHub for the broad approach... See http://github.com/gtalug/backups That repo doesn't include this, as my router's config isn't GTALUG business, but the approach in the scripts in the "scripts" directory should be suggestive. I wee while ago, I retried the WAN link, basically moving the plug from one router to the other, and, a mod to gateway addresses in /etc/network/interfaces later, all now seems copacetic, with no particularly interesting changes made to explain why incoming traffic wasn't happening Saturday but is working now. At any rate, my old Cisco/Linksys WRT310Nv2 will be retiring as consequence. The one interesting area that I probably need to poke at more is that of IPv6 support. The old router didn't have any (evident) support. OpenWRT seems to. I haven't fiddled with the configuration, as I could readily see that breaking things. I had been suppressing IPv6 activity (e.g. - my internal Bind9 /usr/sbin/named instances being run with the "-4" option that suppresses IPv6 usage); perhaps that ought to change now? I'm not quite sure.