I am a newbie with MikroTik but have lots of experience with OpenWrt and ddWrt. It’s now nice to have a supported router with lots of power and functionality.
But I can’t figure out how to create a table of host names that map to some dhcp addresses that I have made static. For instance, I have about 20 VOIP phones in the office that I want to migrate to a new RB2011UAS and I would like for each phone to have a name such as ext100.tik.pbx, ext101.tik.pbx … ext200.tik.pbx (you get the idea). I could do this with openwrt and the ‘ethers’ file and the ‘hosts’ file and ‘dnsmasq’.
How would I implement this with RouterOS. It does not seem obvious to me and I have tried some of the suggestions in this forum with no luck. I am running V 6.6.
Make sure you enter hosts using FQDN like myphone.myNet.local. If you want to ping these IP’s from Windows workstations you’ll need to do an ipconfig /flushdns to clear what you currently have.
So I have the DNS working for local hosts on the LAN with a FQDN. For example I can use a host name to browse or ping such as ext125.hosts.tik where ext125 is a voip phone on the lan.
Now the big BUT. But the dns name resolution is only working on my linux hosts and the router console itself. Windows 7 machines don’t like what I am now leasing to them when it comes to DNS.
I have set the DHCP server to send 192.168.88.1, 8.8.8.8 and 8.8.4.4 as the nameservers (in that order) and the DNS Suffix Search List is showing as “hosts.tik” when i run “IPCONFIG /all” command but don’t know why a ping fails on “ping ext125.hosts.tik”.
So I am closer to a resolution but still not what my existing OpenWRT and DD-WRT with DNSMASQ give me in my soon to be legacy environment.
Because when there’s only one server to query, client can’t ask a wrong one. Previously you had three. And while redundancy is always nice, in this case it didn’t help you, because only one server knew the correct answer. And it doesn’t work in a way that client asks another server if first doesn’t know. It does that only when server does not respond at all. If it’s alive, it sends back clear “sorry, this does not exist” and client has no reason to doubt it and ask elsewhere.
But 192.168.88.1 was first, right? Some sources suggest that Windows don’t care about that and even change used server periodically. I can’t confirm that (read “I never checked”), but in any case, there’s always a chance that one of the other servers gets used and it will break it for you, because they simply don’t have your local data.
But what about redundancy, i.e. safer operation? What if 192.168.88.1 does not work? Right, but isn’t 192.168.88.1 your only gateway to internet anyway? Because if it is and it dies, having another external dns servers on clients won’t help anything. In theory, only dns might fail and the rest could work and then another dns server would help. But it brings back the original problem. So lets hope that does not happen.
This seems like the most direct answer to this question in all the posts I’ve searched, but I can’t see where to implement that. Can you show me how that’s done either in the CLI or web interface?
