Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I've got a site I'm having monitored by pingdom and another service, and both have reported outages. One report specifies the IP address could not be resolved -- however, our DNS is handled by theplanet's DNS servers (ns1.theplanet.com, ns2.theplanet.com) and I can only imagine they have tens of thousands of domains -- I can't imagine this woudl be going on for weeks without it being fixed.

Is there something that could cause the IP address to not resolve other then the DNS servers?

share|improve this question
    
Firewall issues? –  0xC0000022L Mar 8 '11 at 4:15
    
Wait... The IP doesn't resolve or the hostname doesn't resolve? Do you have the exact text of the error? –  Insyte Mar 8 '11 at 4:47
    
it failed to get the IP address for the host. The error pingdom gave was thier own error msg "unknown target" and it has a red error mark next to "Resolve IP" –  ESW Mar 8 '11 at 4:50

1 Answer 1

up vote 4 down vote accepted

If the problem is failure to resolve, then it pretty much has to be a problem with the DNS servers or configuration. It's possible, albeit unlikely, that Pingdom and the unnamed other service have broken caching resolvers. It's also possible, ableit unlikely, that theplanet has broken DNS servers.

It seems more likely that the DNS zone for your domain might be mis-configured, resulting in random failures. Lame delegation from the GTLD servers, for example, can result in intermittent failure. If you post the actual domain, we could tell you quite a bit more.

Some things to check:

dig ns example.com @f.root-servers.net

This will return a list of additional DNS servers that have been delegated to handle DNS for your TLD. Pick one and send the same query to it. For example:

dig ns example.com @f.gtld-servers.net

This should return a list of the servers that handle the DNS for your actual domain. Hopefully it will be ns1.theplanet.com and ns2.theplanet.com.

Next, move on to sending an A request to each of the listed servers:

dig a example.com @ns1.theplanet.com
dig a example.com @ns2.theplanet.com

(Assuming that the GTLD server listed those servers.)

They should both return the same results, differing only in minor details like TTL.

share|improve this answer
    
Thanks for the information -- took all the steps outlined above; the gtld servers returned the proper DNS servers ( ns1.theplanet.net & ns2... ) and the dig a requests to theplanet returned the proper IP for our amazon EC2. Its a very intermittent problem, happening every few days, or sometimes multiple times in a single day. The server is rarely unreachable for more then 2-3 minutes. –  ESW Mar 8 '11 at 15:29
    
Assuming the error you're seeing is correct (by no means a given) then it pretty much has to be either a failure in the monitoring app or a failure at theplanet's DNS servers. All I can suggest is that you set up your own monitor on a remote host somewhere. Have it run dig, both using the local resolver and directly against theplanet's mail servers, every few minutes and log it to a file. –  Insyte Mar 9 '11 at 8:42
    
I think it must the thePlanet then. We're using multiple monitoring services although I find the second sketchy at best. When pingdom gets a failure it retries it right away from a 2nd server in another physical location, and overall seems to use servers all over the USA/Europe. Now if I can only manage to convince thePlanet there's a problem. –  ESW Mar 9 '11 at 16:19
    
You'll find that much easier if you can back it up with data. I strongly recommend you start logging your own DNS queries as described above. –  Insyte Mar 9 '11 at 19:15

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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