thanks in advance.

i have a DNS problem.
i have one of my servers in denver on yipes. recently, yipes had 29 straight hours of downtime at the level3 interconnect there. (needless to say i no longer love yipes and you can bet the server will be moved, which will be in 3 weeks, which is not soon enough in my opinion)


now, i have 3 personal domains on this box.
1. my main domain (on name based IP) with 2 NS setup (of course on their own IPs).
2. domain that is on the main name-based IP.
3. domain that has it's own domain.

all 3 domains use the main domain's nameservers to resolve.

now, after the lines have come back up, ONLY my main domain (1)is resolving and working correctly. the other 2 (2 and 3) are no longer resolving.

none of my settings have changed at all. the IPs still resolve to the server perfectly fine as does my nameservers.

the server had amazing uptime and all 3 domains were working perfectly before this downtime.

my question, could the downtime be the direct cause of this. caching somewhere somehow?

i'm effing lost with this and now going on 48 straight hours of downtime for 2 of my domains.

i have double checked all my DNS settings for the domains in PLESK (none of which i changed in the first place) and they're all fine. i have double checked my hostnames and nameservers at godaddy (where my domains are registered) and they seem all fine (none of the settings there were changed either).

thanks for any ideas you may have.

Not enough info for this fool to diagnose but may be you can try a restart of BIND?

It can be because of cacehing, since your nameserver was down, any request to your nameserver was cached as unreachble during the down time.

thanks both of you!
we have tried a restart, still no go.


the chaching is all i can think of right now. driving me crazy. any idea how often the caches are refreshed? (i have the ttl set pretty low in my SOA)

Originally posted by Ahmad
It can be because of cacehing, since your nameserver was down, any request to your nameserver was cached as unreachble during the down time.

If you list the domains, we might be able to help you locate the problem. Negative caching shouldn't last 48 hours.

ah, and i had just PMd you toasty.

but, i THINK i have figured it out. somewhat. did a dns check and find problems with my SOA.

i will be working on it for the next few minutes.


EDIT TIME
nope, now i'm still hoping it's the reverse cache caca. grr.

Like Toasty said, the domains would be useful, dig can be a powerful ally in the land of DNS

thank you all again.

toasty gave me an idea over PM which fixed it all up.


case closed. thanks again.

Can you let us know wat was the the issue??


-ananthan