Re: masters, slaves, and when I make changes
On 27.10.08 20:19, Scott Haneda wrote:[color=blue]
> Hello, I hope this should be fairly simple, most of this is just me
> looking to understand how a certain process works.
> I have a primary NS where I add in new domains, delete old ones, and
> of course, update existing ones. My colocation provider has several
> NS's, but I only use one as a secondary, and only list one as a
> secondary in my NS records.[/color]
Don't do that, unless colo approves it's ok. I'm angry at curstomers that
list servers as they decide, because that may break some things.
If colo provides several nameservers, ask them what to use.
If you'll have more nameservers, you'll have lower chance lookups to fail.
> I am guessing, the colo starts with ns0.colo.com, and each of the ns1
> through ns3 are slaves.[/color]
We have 4 real nameservers, they are all masters for our zones (they are
distributed non-DNS way), colo may be doing the same.
> What determines to them, when the ns1.colo.com, through ns3.colo.com
> will pick up on the new data in ns0.colo.com?[/color]
their configuration, of course.
> When I make a change, it seems to take a very long time for
> ns1.colo.com to pick it up, but I can see that ns0.colo.com has it. I
> send in the notify, and ns0.colo.com grabs relatively quickly. The
> trouble I have, is that their secondary does not seem to get the new
> data for far too long.[/color]
It's just because you only use one ns in your NS records.
Matus UHLAR - fantomas, [email]firstname.lastname@example.org[/email] ; [url]http://www.fantomas.sk/[/url]
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
42.7 percent of all statistics are made up on the spot.