DNS server stop working from time to time

RouterOS General
1

Hello,

I’ve a strange problem with the Mikrotik DNS server of my CCR 1009. The setup is pretty classic :

[admin@MikroTik] > /ip dns print
servers: 80.67.169.12,8.8.8.8
dynamic-servers:
allow-remote-requests: yes
max-udp-packet-size: 4096
query-server-timeout: 2s
query-total-timeout: 4s
max-concurrent-queries: 300
max-concurrent-tcp-sessions: 80
cache-size: 4096KiB
cache-max-ttl: 2d
cache-used: 63KiB

So, all my pc are using the Mikrotik ip address as DNS server, and it’s working fine. For a while. Then, all of the sudden, it’s not working anymore. The DNS doesn’t resole anything, and, if I try a nslookup from one of my windows machine, it’s timing out, like the DNS “service” of the mikrotik is crashed. So, 2 “solutions”. Either I reboot the router, or I modify something in the config, press apply, and boom, dns is up again.

I’ve this problem since 6.38. I’m using 6.39rcs now, and the problem is still here. I can be good for weeks, or hours, it’s really seems random.

I saw a similar post here from 2013 : http://forum.mikrotik.com/t/dns-server-stops-working/72231/1 and no solution except using another dns server.

Thx you.

2

Report it to support as a bug.

3

Hi.

Ok, what is the “official” way to submit a bug report ? I can’t seem to find it, sorry :confused:

Thx.

4

I usually do it via email. Refer here: https://mikrotik.com/support

5

Yeah, I did it yesterday :slight_smile: Thx :slight_smile:

6

Are you runsning any interfaces with dhcp client on them there is a similar issue reported by a number of users where by dhcp client dns causes the issue. they found turning off that dns resolved the issue. I am using the ccr1009 for the past 8 months as dns without issue.

7

I do.

But I found the “true” problem in my case.

I mailed them this : (sorry for my bad english)

"Additional info :

In fact, it seems that Mikrotik sometime doesn’t try all the dns servers in the list, but only the last one. And if the last one is down, then, it doesn’t try the first one, so DNS failed.

[admin@MikroTik] > /ip dns print
servers: 80.67.169.40,80.67.169.12
dynamic-servers:
allow-remote-requests: yes
max-udp-packet-size: 4096
query-server-timeout: 2s
query-total-timeout: 4s
max-concurrent-queries: 300
max-concurrent-tcp-sessions: 80
cache-size: 4096KiB
cache-max-ttl: 2d
cache-used: 36KiB

In this case, when I’m taping this email, 80.67.169.12 is down. But 169.40 is up. BUT, only 169.12 is “used”, so everything failed.

When I switch the 2 servers orders :

[admin@MikroTik] > /ip dns print
servers: 80.67.169.12,80.67.169.40
dynamic-servers:
allow-remote-requests: yes
max-udp-packet-size: 4096
query-server-timeout: 2s
query-total-timeout: 4s
max-concurrent-queries: 300
max-concurrent-tcp-sessions: 80
cache-size: 4096KiB
cache-max-ttl: 2d
cache-used: 30KiB

It’s working again, because 169.40 is used, and not 169.12. Which is the correct thing, but, it seems it doesn’t even try the first servers, only hitting the last one, even if it doesn’t work.

Even when the 2 servers are working, often only the last one is used. So their is not real redundancy here… Joined the supout.rif file this time.

Thx !"

8

Hi
I have the same issue after upgrading our CRS1036 to v6.48.1. The internet on router is still active and I can ping to 8.8.8.8 or 8.8.4.4 but end-client does not.
So I have to change DNS server to work, It quite strange after awhile the new DNS server also can not reached on client side.
Reboot will make it work again.

9

Similar here with RB4011 v v6.48.3. Sometimes even without reboot it starts working again after some time. My solution is to put a backup external DNS server to clients (via DHCP, primary is MT DNS, secondary is external).