Re: Re updating the "DNS Debugging Tools" Doc
Michael Sondow <[email protected]> Fri, 02 Apr 1999 10:10:39 -0500
| Newsgroups | gmane.ietf.fyiup |
|---|---|
| Message-ID | <[email protected]> |
Gary Kessler a écrit: > Check out the DNS Resources Directory at http://www.dns.net/dnsrd/. I did, and found it to be an excellent resource. I've sent the URL to my ISP (although, who knows if he'll use it?). > I also have a paper titled "Setting Up Your Own DNS" at http://www.hill.com/library/dns.html; it's a little old right now but I'd be happy to update it if you'd like. I'll check that out today. Thanks. And if you'd like to update it, that would be very nice, too. The FYIs and RFCs on setting up the DNS records, and on the DNS tools, don't cover enough. The problem with them is that they don't give enough examples of mis-configured records and how to fix them (short of starting all over again). My ISP's zone files are in a terrible state. This seems to be surprisingly common among small-ish ISPs. For example, when and how to use CNAME is not very clearly explained in the BIND books. Same goes for the MX records. My ISP used to have all sorts of bad CNAME records in the zone files for me and for a lot of other customers, until an engineer at the Mexico NIC did me the favor of going through my ISP's records and making a list of all the bad ones. But even he missed a few. Another big problem now is Y2K compliance in BIND. These little ISPs who set up their accounts and records four or five years ago don't realize how many non-compliant strings they've got in their files. I convinced my ISP to search for "9*" in the files, and we found seven instances of it without the "19" in his own records, with three in the NS record! In other words, any further information will be most gratefully appreciated. Until the day when there's some sort of external control of DNS records (an automatic, third-party check-up of the records of everyone running a name server?), the only thing that can be done is to make information and tools available.