Domains not reachable, zone files expired

  • Hello,


    Today all of my older domains weren't reachable because the zone files were expired. Changing the serials of the zone files solved the problem. Is there a way to disable the expiring of zone files ?


    Gretings,
    T0mcat

  • Hello,


    I'm using Bind9 and a second nameserver. I never had any problems before upgrading to 1.1.5. Both servers were working for 3 years now (with ispcp and then i.mscp).

  • Thanls ciscllc for your help. Here is my anonymized zone file:



    Greetings,
    T0mcat

  • Mine looks nearly the same.
    And this one is way older than 2 weeks. But at an other Server I have seen a rebuild of the files 2 weeks ago (16 april 2014 as we have the 2nd may today...) Maybe @Nuxwin can tell us what file is doing that (as I have not done that myself)


    EDIT: 16.04.2014 I made an update! Thats why the files have changed... Sorry



    Regards

  • Hello ;


    All this is managed through the expiry and refresh fields of the SOA record.


    If the secondary nameserver is unable to update its zone by contacting the primary DNS server in the interval time as stated by the expiry field, the zone will be considered as expired, meaning that your services (web, smtp..) will become unreachable in few time. In default configuration as provided by iMSCP, the slave zones are refreshed every 3 hours normally. This should avoid the expiry time to be reached. Because the expiry time is reseted on every refresh.


    To solve your problem here, you must ensure that your secondary DNS server is able to refresh its zone file by contacting the primary DNS server. If it's not the case, you should provide us the log and open a ticket.



    Ref: http://www.zytrax.com/books/dns/ch8/soa.html

    badge.php?id=1239063037&bid=2518&key=1747635596&format=png&z=547451206

  • @ciscllc the pingdom check doesn't provide any errors. The intodns.com/domain.tld also doesn't show any errors.


    @Nuxwin I'm keeping an eye on the secondary nameserver, but for now, the problem seems to have gone since a week.