Domains not resolving in Bind9, flush cache, clears issue.

Brook Harty harty at ironwolve.com
Tue Apr 13 17:23:45 UTC 2004


phn at icke-reklam.ipsec.nu wrote in message news:<c5fulg$2rs4$1 at sf1.isc.org>...
> Brook Harty <harty at ironwolve.com> wrote:
> > phn at icke-reklam.ipsec.nu wrote in message news:<c58b8g$1p46$1 at sf1.isc.o=
>  rg>...
> >> Brook Harty <harty at ironwolve.com> wrote:
> >> > Not sure if this is a bug, but I'm having wierd cache type issues on
> >> > Bind9.
>  =20
> >> > I upgraded all our bind solaris8 servers from bind 8.2.3 to Bind9.2.=
>  2
> >> > (then 9.2.3). My 2 DNS servers that sit on the Internet after some
> >> > time (18+ hours) starts having issues resolving a couple extrenal
> >> > domains. Flush cache, and they would work again.  I only have 2 that=
>  I
> >> > know of, it could be more.
>  =20
> >> > When I would try to resolve these external domains having issues, it
> >> > just reports
> >> > domainname IN A =20
>  =20
> >> > Thats it, no IP.=20
>  =20
> >> > I did a trace before and after flushing the cache, also a db dump
> >> > before and after flushing cache. Theres nothing that looks out of
> >> > place in the dump, the trace shows it not resolving.
>  =20
> >> > I reverted to bind8, and its resolving these domains again. Can't fi=
>  nd
> >> > a common problem with the remote domains, and why Bind9 would just
> >> > fail after some hours.
>  =20
> >> > Where should I go from here? I never submitted a bug for bind. Reall=
>  y
> >> > would like to use Bind9 for my Internet facing DNS servers.
>  =20
> >> > Any ideas appreciated.
> >>=20
> >> Tell us what domains you have problems with.
>  
> > One of the domains is www.icee.com, domain isn't lame.  BTW, I'm back
>                         ^^^^^^^^^^^
> This is not a domain, it's a CNAME to a domain.
> 
> And it's broken shure. One of the two servers=20
> ( mailscan.gourmettwists.net ) delegated to is missing from the
> zonefile=20
> 
> 
> > to 8.x, and not a single issue with domains not resolving. Here is a
> > dig, cache entries, and a trace all ran before and after flushing
> > cache.
> bind8 trusts gluerecords while bind9=20
> 
> > Why would it work for awhile, then fail?
> 
> Because the glue record will time out.
> 
> 
> --=20
> Peter H=E5kanson        =20
>         IPSec  Sverige      ( At Gothenburg Riverside )
>            Sorry about my e-mail address, but i'm trying to keep spam out=
> ,
> 	   remove "icke-reklam" if you feel for mailing me. Thanx.


Yup, now I see all the domains having issues are glue record issues.
Must be a common problem when upgrading to Bind9.  How do larger ISP's
avoid this problem, just stick with bind8?

BTW, Thanks for the help.


More information about the bind-users mailing list