3

(All IPs and names have been censored, hopefully)

This morning we had some connectivity problems in one of our servers. It turns out that changing our connection configuration to be IP-based (not preferable) solves this, but surprisingly DNS did not completely fail.

What happens is that the DNS sometimes (within a range of seconds) manages to translate the given url to the corresponding IP and sometimes doesn't (unknown host).

What could be causing such behavior? What steps can I take to make sure this is handled well? Other

badumtsss@02-02-02-02:~$ ping dingding.badumtsss.com

ping: unknown host dingding.badumtsss.com

badumtsss@02-02-02-02:~$ ping dingding.badumtsss.com PING dingding.badumtsss.com (01.01.01.01) 56(84) bytes of data.

64 bytes from ec2-01-01-01-01.eu-west-1.compute.amazonaws.com (01.01.01.01): icmp_req=1 ttl=43 time=18.7 ms 64 bytes from ec2-01-01-01-01.eu-west-1.compute.amazonaws.com (01.01.01.01): icmp_req=2 ttl=43 time=18.5 ms --- dingding.badumtsss.com ping statistics --- 8 packets transmitted, 8 received, 0% packet loss, time 7010ms rtt min/avg/max/mdev = 18.523/18.721/18.815/0.147 ms

badumtsss@02-02-02-02:~$ ping dingding.badumtsss.com

ping: unknown host dingding.badumtsss.com

badumtsss@02-02-02-02:~$ ping dingding.badumtsss.com PING dingding.badumtsss.com (01.01.01.01) 56(84) bytes of data. 64 bytes from ec2-01-01-01-01.eu-west-1.compute.amazonaws.com (01.01.01.01): icmp_req=1 ttl=43 time=18.5 ms 64 bytes from ec2-01-01-01-01.eu-west-1.compute.amazonaws.com (01.01.01.01): icmp_req=2 ttl=43 time=18.7 ms ^C

--- dingding.badumtsss.com ping statistics --- 2 packets transmitted, 2 received, 0% packet loss, time 1001ms rtt min/avg/max/mdev = 18.586/18.681/18.777/0.166 ms badumtsss@02-02-02-02:~$ ping dingding.badumtsss.com PING dingding.badumtsss.com (01.01.01.01) 56(84) bytes of data. 64 bytes from ec2-01-01-01-01.eu-west-1.compute.amazonaws.com (01.01.01.01): icmp_req=1 ttl=43 time=18.6 ms 64 bytes from ec2-01-01-01-01.eu-west-1.compute.amazonaws.com (01.01.01.01): icmp_req=2 ttl=43 time=18.6 ms ^C

--- dingding.badumtsss.com ping statistics --- 2 packets transmitted, 2 received, 0% packet loss, time 1001ms rtt min/avg/max/mdev = 18.612/18.616/18.621/0.136 ms badumtsss@02-02-02-02:~$ ping dingding.badumtsss.com ping: unknown host dingding.badumtsss.com

  • Any connectivity problem, whether client-side or server side. – MariusMatutiae Oct 17 '13 at 7:38
  • IPs work flawlessly so far – Reut Sharabani Oct 17 '13 at 8:19
  • Then the problem is in the connection to your DNSs. Have you tried changing them? – MariusMatutiae Oct 17 '13 at 8:21
  • Not yet, but the same DNS servers work on other servers. I'll try and do that, but ubuntu changed it's old DNS-changing procedure. I thought this may be a common problem that I don't know of. – Reut Sharabani Oct 17 '13 at 8:24
1

Look at the list of DNS servers in /etc/resolv.conf and try to perform the queries directly using e.g. dig and see whether you get the correct answer and how long it takes. You can use tcpdump -vv or a tool like Wireshark to see what queries are used by the applications.

dig some-domain A @name-server

Theoretically you could improve the situation by using a local caching nameserver (e.g. unbound, dnsmasq or even bind), but that cover the real problem at best.

0

Are you perhaps using ec2 for these queries, using the AWS resolver. If so, open a support case and have them look at the resolver/host combination.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.