D. J. Bernstein
Internet publication

Management of the namedroppers mailing list


The DNS protocol is covered by various IETF specifications. Unfortunately, obeying those specifications is not sufficient to ensure interoperability with BIND, in part because the specifications are ambiguous or otherwise flawed, and in part because BIND violates the specifications in many ways.

These facts have hurt competition, and contributed to BIND's market share, at the expense of the users. For example, one site using lbnamed, a special-purpose DNS implementation, has had interoperability problems with BIND, and has been planning to abandon lbnamed in favor of BIND, even though this means giving up some useful features.

In late 1999, after yet another BIND security hole was announced, I wrote a free BIND replacement. Interoperability among DNS implementations is, of course, essential. I found the IETF specifications horribly inadequate.

The namedroppers mailing list

IETF carries out its DNS protocol standardization activities within the DNSEXT working group. The DNSEXT mailing list is namedroppers@internic.net, also known as comp.protocols.dns.std.

``Within the scope of this WG are protocol issues, including message formats, message handling, and data formats,'' the DNSEXT charter says. Several specific issues have been identified as work items, but other DNS protocol issues remain clearly within the charter. In particular, namedroppers is obviously the right forum for implementors to discuss current and future DNS interoperability problems.

Unfortunately, namedroppers is being run in a way that slows down, and sometimes prevents, public communication among DNS implementors.

Messages to namedroppers are not forwarded directly to subscribers. They are first sent to Randy Bush. They wait for Bush's review. Bush discards, edits, or misdirects messages that he doesn't like, and passes along what's left.

Here are some specific examples. Many of these incidents involved opsmail.internic.net, which used some painfully slow, creaky, obsolete software to distribute messages to subscribers.

IETF review

1998-12-03: I sent a message to the IESG and a few other people, describing two of the incidents listed above. I didn't know who was responsible. I asked whether the IESG was aware that this sort of thing was happening. I also asked whether all previous subscribers were aware of it.

``There was some discussion some time ago when spam started to be a problem,'' Robert Elz said. But what actually happened, according to namedroppers archives, was that Bush responded to some spam in 1996-04 by unilaterally restricting the mailing list. He said, at the time, that he would reject ``just the spam and administrivia.'' There is no evidence that the IETF DNS working groups ever approved of Bush's censorship of on-topic messages.

``Randy Bush and Mark Kosters jointly moderate namedroppers, just to keep spam out,'' BIND maintainer Paul Vixie said. That's consistent with what Bush claimed on the mailing list in 1996, but it's not true. What Bush did was not ``just to keep spam out,'' but to actively and deliberately bias the mailing list discussions.

I requested (twice) that the IESG stop using censored mailing lists for standardization activities. Scott Bradner told me that the IESG had discussed my request, and that I should be receiving a formal response from the IETF chair. I never received any such response. However, I decided that I had made myself sufficiently clear, and that I wouldn't press the issue unless I saw evidence of continued misbehavior by Bush.

1999-12-19: Unfortunately, the incidents continued. I sent another complaint to the IESG.

``The IESG recognizes that the moderating of IETF mailing lists is a sometimes necessary and appropriate tool to help manage WG activities,'' Narten said, ignoring the facts of the case. He told me to complain to the two WG chairs: Olafur Gudmundsson and Bush.

2000-01-08: I sent a complaint to Gudmundsson, with a copy to Bush. Gudmundsson sent me a useless response two days later:

Most revealing was Gudmundsson's final remark that he had an ``agenda'' for DNSEXT and would not tolerate ``disruptions.'' Evidently he thinks that fundamental interoperability questions and suggestions from a new implementor, all clearly within the DNSEXT charter, are ``disruptions.''

I explained my interoperability concerns to Gudmundsson. I asked him to obey RFC 2418, section 3.2, which allows working groups to consider mailing-list controls but requires IESG approval of those controls. I asked him to reveal Bush's activities to the working group.

In the next five days, Gudmundsson sent several messages to namedroppers within his ``agenda,'' but he didn't respond to my message.

2000-01-15: I sent a complaint to the area directors, Erik Nordmark and Narten. I didn't receive a response.

2000-01-20: Gudmundsson announced that namedroppers would be moving to ops.ietf.org, another name for Bush's machine psg.com, in a week. (psg.com is slow by modern standards, but not as slow and hopefully not as fragile as opsmail.internic.net.) Gudmundsson still didn't mention this web page.

I noticed that the obsolete software on opsmail had been configured with a namedroppers-outgoing address that would allow anyone to send a message directly to the subscribers, without Bush having a chance to censor the message. It was clear that this opportunity wouldn't last. I sent a message to the subscribers, pointing out this web page. The responses included evidence of further misbehavior by Bush.

Meanwhile, I sent a complaint to the IESG. ``Some of the incidents were caused or exacerbated by software problems, which the WG chairs are finally attempting to fix,'' I said. ``However, most of the incidents were selectively and deliberately caused by one of the WG chairs. The other WG chair has made clear that he is not going to fix those problems.'' I explained that I had already contacted the area directors without receiving a response.

``Your complaint was received by the Internet ADs over the weekend,'' Narten wrote. ``We are currently evaluating the complaint and developing a response. You can expect to get a more detailed response by sometime next week.''

``What's taking you so long?'' I asked. ``Why don't you say what you think right now? This isn't some tricky technical issue. It's simple abuse of power by an IESG agent.'' I also asked whether Bush's behavior had already been approved by the IESG. There was no response to these questions.

2000-01-26: Narten (apparently also speaking for Nordmark) sent me a useless response, which arrived from opsmail 17 hours later:

Narten didn't explain why this content-free message had taken him so long to write.

I sent yet another complaint to the IESG. I received no response.

2000-02-02: I sent a complaint to the IAB. IAB chair Brian Carpenter refused to consider my complaint ``until the IESG has responded.'' He added that ``responses to appeals generally take one to two months to prepare.''

I explained that the IESG had already had ample opportunity to respond, and that Bush was blatantly violating RFC 2418 if the IESG had not already approved his censorship. Carpenter again refused to consider my complaint.

2000-02-04: I sent a complaint to the ISOC board of trustees. ``Executive summary: The IAB/IESG/IETF standardization procedures, as written and as used in practice, fall far short of the requirements of United States antitrust law,'' I wrote. I gave an example of misbehavior by Bush, an example of a misbehavior by John Klensin in another working group, and pointers to details of other incidents. ``My complaint is not merely with the behavior of these people, but also with the procedures that have allowed such behavior,'' I wrote. I explained some of the FTC requirements on standards organizations.

After ISOC president Don Heath sent me a useless response, I sent a second message to the ISOC board of trustees. ``How many years do I have to wait before you engage in the review required by RFC 2026?'' I asked.

Heath did not respond. RFC 2026, section 6.5.3, specifically requires that Heath ``advise the petitioner of the expected duration of the Trustees' review'' within two weeks; Heath did not do this. In fact, it appears that the ISOC board of trustees is completely ignoring the RFC 2026 requirements to review this situation and report to the IETF. ``The ISOC Board is a nonentity,'' one board member wrote to me privately. ``It won't fix anything until it is forced to.''

2000-10-11: IESG chair Fred Baker sent a response to the complaint that I had sent IESG eight months earlier:

Baker also announced a policy on mailing list censorship:

2000-11-15: I sent another complaint to the IAB.

2000-12-13: IAB member John Klensin sent me a useless ``initial response'', attempting to mischaracterize the substance of my complaint. I sent a more detailed complaint to the IAB.

2001-02-26: Tony Hain, speaking for the IAB, sent me a useless response.

According to RFC 2026, it is up to the IAB to decide ``whether or not the Internet standards procedures have been followed.'' But Hain explicitly refused to consider any issue other than ``the manner of the handling of the appeal by the IESG,'' which is a meta-issue of much less importance that Bush's blatant violations of RFC 2418.

RFC 2026 specifically requires that the IAB ``report to the IETF on the outcome of its review.'' Hain sent his message to me, the IAB, and the IESG, but not to the IETF.