- db-wg - mailman.ripe.net

RE: Default behaviour for referred domain queries
by Mike Norris 06 Oct '98

06 Oct '98

In coincidence with your input, the db-wg also found that whenever the server shows data from a remote server instead of the local data in response to a query, the response should be preceded by a notice clearly informing the user that the information being displayed is coming from another server that contains the authoritative data. Right. It was also noted that the original proposal did not provide a way to look at the "local" object if it contained a refer attribute. In order to avoid potential confusion from a mode of operation where two objects are displayed, it would be better to provide a whois flag (-R was agreed) that would disable the referral and show the local object. Bear in mind that there could be multiple referrals involved i.e. the process is recursive. Suppose the query is for a third-level domain, that the second-level domain is referred from the RIPE DB to a TLD registry server, and that the target domain is referred onward from the TLD server to an ISP's whois server. Now what will the -R flag deliver - the local object in the RIPE DB? That'll give you the coordinates of the TLD server, which you again query with -R to get the ISP 'local' object. Regards. Mike

1 0

Default behaviour for referred domain queries
by Niall O'Reilly 05 Oct '98

05 Oct '98

I'ld like to give input as requested on this topic. For the TLD people who may have missed the DB session, here is first a brief description of how the referral feature works. I don't know the details at this stage. When a request is received by the "central" server for a domain object marked (locally) for referral, the query is referred to the authoritative server and the data from the authoritative server is returned to the client. If the domain object requested is not found on the "central" server, the specified domain name is truncated (successively, if necessary ?) until a higher-level domain is found to be marked for referral, and the query is then referred to the authoritative server for that (higher-level) domain. The following suggestions seem to be almost agreed, but need to be written down. By default, the response returned to the client for a referred query should include the following elements: Alert that referral has been used Response from authoritative server I expect that the authoritative response will include a copyright and/or AUP banner from the authoritative server, but this is the responsibility of the authoritative registry, and not of the registry providing the referring server. I expect that a 'verbose mode' will provide a response including the following elements: Response from referring server as if no referral was used, including referring registry's standard (C)/AUP banner Alert that referral is spin effect for the domain in question Response from authoritative server Whichever of the modes described above is used, it should be possible to specify whether recursive lookup and display of related person/role (and in future, nameserver) objects is to be done. IHTH, Niall

2 2

Re: NIC handle writeup
by Wilfried Woeber, UniVie/ACOnet 02 Oct '98

02 Oct '98

[ Apologies for prematurely pressing send. It's Friday afternoon :-( ] => "%" still has mail-related connotations though, although far less so than => "@". => => How about ":"? JA39:APNIC.NET doesn't look like an e-mail address or a DNS => name. It does however look slightly remotely like a URL ;) = = ":" not only has URL significance, its also used w/ IPv6 in the = address. IMHO a poor choice. I don't think that we're going to find any character which isn't badly overloaded already... => Or have I misunderstood the point you were making? = = I think so. Back up a bit and take the 10,000m view. Yup: why do we need a special character at all? Grab a handle, which simply must not contain the "." character, and extend that with the DNS-part as proposed. Done. And it is known to work (DNS SOA record) My good old WW144 would then simply become WW144.INTERNIC.NET. = There are three classes of Internet intangibles that = can be delegated, ASN's, Prefixes, Port numbers. (there = are others, but these will do for this purpose) = = The recipiants of these delegations can be people (admin-c/tech-c), = hosts (HST), organized groups (mntner & role accounts) While this is probably off-topic, I suppose you are mixing organisations and contact information here... = I could make the case that the mntner:, admin-c:, tech-c:, = mnt-by:, nic-hdl:, and the HST (C8-HST) are all, in a very = real sense, NIC HANDLEs. They specify a relationship between = Internet resources and those who are authorized to manage the = delegation relationship. I have to disagree. - We do have mechanisms to describe the privilege(s) for maintaining objects. - And we do have mechanism to get in touch with persons (or helpdesks or NOCs) for operational purposes. = Here, I expect that the handle: = = WM110 = WM110-NSI = WM110-ARIN = = all reflect me, first under the SRI-NIC, then Internic/NSI, then = ARIN. I would hope the design of NIC-handles would allow migration = of data between registries as well as allow for a deeply nested = registration heirarchy. What we've learned from past (failed) attempts: - don't try to mix the issue generating globally unique identifiers with - the issue of data exchange between registries As soon as we've sorted out the first one, we can try to work on the second one. Having the first one done is not going to interfere with solving the other issue, I suppose :-) Wilfried. -------------------------------------------------------------------------- Wilfried Woeber : e-mail: Woeber(a)CC.UniVie.ac.at Computer Center - ACOnet : Tel: +43 1 4277 - 140 33 Vienna University : Fax: +43 1 4277 - 9 140 Universitaetsstrasse 7 : RIPE-DB (&NIC) Handle: WW144 A-1010 Vienna, Austria, Europe : PGP public key ID 0xF0ACB369 --------------------------------------------------------------------------

1 0

Re: NIC handle writeup
by Wilfried Woeber, UniVie/ACOnet 02 Oct '98

02 Oct '98

From: MX%"bmanning(a)ISI.EDU" 25-SEP-1998 20:56:00.16 To: MX%"jabley(a)clear.co.nz" CC: MX%"db-wg(a)ripe.net" Subj: Re: NIC handle writeup From: bmanning(a)ISI.EDU Subject: Re: NIC handle writeup To: jabley(a)clear.co.nz (Joe Abley) Date: Fri, 25 Sep 1998 11:50:08 -0700 (PDT) CC: db-wg(a)ripe.net > "%" still has mail-related connotations though, although far less so than > "@". > > How about ":"? JA39:APNIC.NET doesn't look like an e-mail address or a DNS > name. It does however look slightly remotely like a URL ;) ":" not only has URL significance, its also used w/ IPv6 in the address. IMHO a poor choice. > > > The only thing that > > > really matters is if we can find the data and that it improves the > > > current situation. > > > > Thats two things. While desirable, the thing that I thought > > NIC handles did was to identify an entity that was responsible > > for some delegated level of responsibility. > > > > This could be a host, person, role, or organization that > > accepted the responsibility over some delegated Internet > > resources (numbers & lables). > > In ripe-181 the habit seems to have been to put contact fields (tech-c, > admin-c) in every record where contact was appropriate, so contact details > via NIC handles are immediately available. In other words, the drill down > path of (for example) inetnum -> netname -> admin-c/tech-c isn't necessary > since there are direct relationships inetnum -> admin-c and inetnum -> tech-c. > > Has this changed significantly in RPSL, so that it is now important to have > netname-type fields which have registry-internal and registry-external > representation? > > Or have I misunderstood the point you were making? I think so. Back up a bit and take the 10,000m view. There are three classes of Internet intangibles that can be delegated, ASN's, Prefixes, Port numbers. (there are others, but these will do for this purpose) The recipiants of these delegations can be people (admin-c/tech-c), hosts (HST), organized groups (mntner & role accounts) I could make the case that the mntner:, admin-c:, tech-c:, mnt-by:, nic-hdl:, and the HST (C8-HST) are all, in a very real sense, NIC HANDLEs. They specify a relationship between Internet resources and those who are authorized to manage the delegation relationship. > > So each of these things would > > have an unambigious lable that could be tied to an assignment > > of responsibility. Those lables should be consistant, regardless > > of where they are listed. They should also have some token that > > indicates the listing origin and may include a method for > > determining the level of trust on the listing. Here, I expect that the handle: WM110 WM110-NSI WM110-ARIN all reflect me, first under the SRI-NIC, then Internic/NSI, then ARIN. I would hope the design of NIC-handles would allow migration of data between registries as well as allow for a deeply nested registration heirarchy. --bill -------------------------------------------------------------------------------- Return-Path: <owner-db-wg(a)ripe.net> Received: from ns2.univie.ac.at by access.cc.univie.ac.at (MX V4.1 VAX) with SMTP; Fri, 25 Sep 1998 20:55:55 MET-DST Received: from postman.ripe.net (postman.ripe.net [193.0.0.199]) by ns2.univie.ac.at (8.8.8/8.8.8) with SMTP id UAA56164 for <woeber(a)cc.univie.ac.at>; Fri, 25 Sep 1998 20:55:54 +0200 Received: (qmail 918 invoked by alias); 25 Sep 1998 18:55:23 -0000 Delivered-To: lists-db-wg-out(a)lists.ripe.net Received: (qmail 916 invoked by uid 66); 25 Sep 1998 18:55:23 -0000 Posted-Date: Fri, 25 Sep 1998 11:50:08 -0700 (PDT) Message-ID: <199809251850.AA08047(a)zed.isi.edu> In-Reply-To: <19980925102909.B15327(a)clear.co.nz> from "Joe Abley" at Sep 25, 98 10:29:09 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-db-wg(a)ripe.net Precedence: bulk

1 0

Re: NIC handle writeup
by Havard.Eidnes＠runit.sintef.no 25 Sep '98

25 Sep '98

> I think that there will be problems with "@" as the > seperator. It's too much like a viable email address. I > applaude the use of the domain name on the right and a > domain specific unique handle on the left, the concern is > the seperator. Howabout something like "%"? I fully agree. Another option could be ":". Then there is the already established practice of using "-" (as in RIPE's case), I don't see what's so terribly wrong with that. Making NIC handles into something which could be mistaken for an e-mail address just leads to confusion. (No, I don't think it's such a great idea to provide automatic mail forwarding service by the registry.) I also don't think it's such a great idea to use domain names to identify the origin of the data; I think a separate name space should be used instead, e.g. one which could provide more permancence. (As Bill mentioned, what do you do when a registry closes down and its corresponding domain name is subsequently removed?) - Håvard

3 2

Slides from RPS Authorization and Replication Talks
by Cengiz Alaettinoglu 24 Sep '98

24 Sep '98

Hi, As some of you already requested, my slides, both on authorization and IRR distribution, are available at: http://www.isi.edu/~cengiz/talks/ RPSL tutorial slides are available at: http://www.isi.edu/ra/rps/training/ Regards. Cengiz -- Cengiz Alaettinoglu Information Sciences Institute http://www.isi.edu/~cengiz University of Southern California

1 0

Re: Cc: in <auto-dbm> replies
by Steven Bakker 14 Sep '98

14 Sep '98

>>>>> On Fri, 11 Sep 1998, "j" == johnpc(a)xs4all.net wrote: j> The return address from auto-dbm is already something else, not auto- j> dbm itself (it's not nobody, it is in fact a human whose "d" key must j> get more sore than the fire button in a cheap shoot-em-up game from j> all the bounces). (You might've noticed by actually looking at an j> auto-dbm reply! :) j> However, if you start copying Cc: lines, you're in trouble. Think j> about it a little and you'll figure it out. Like: To: johnny(a)rotten.net Cc: auto-dbm(a)ripe.net I agree that replying to Cc: is not a good idea, I'd prefer to use Reply-To:, but note that even then it's still possible to "sewiously scwew up", although accidental screw ups are less likely. However, since the From:- address on a auto-DBM reply is always the same, can't you filter out:- "To: auto-dbm" & "From: ripe-dbm" -> bitbucket Or am I missing something extremely obvious here? Like, does the "ripe-dbm" account ever need to send update messages to "auto-dbm"? Furthermore, the classic way of detecting mail loops is by looking at the "Received:" headers. Your MTA should be able to handle that all by itself, no need to program the RIPE DB for it... Of course, none of the above would protect you from forwarders that strip/modify headers... :-( Cheers, Steven

1 0

whois output and the "changed" line.
by RIPE Database Administration 11 Sep '98

11 Sep '98

Dear Colleagues, Recently, the output of our whois server was modified; the "changed" information is not shown by default. To see it, the new "-c" flag must be used. This requires using the new RIPE whois client. The "changed" information was removed from the default output of whois after a discussion on the mailing list of the RIPE Database Working Group. This discussion was about the funtionality of the "changed" line. A long-term replacement for it (i.e. timestamping of updates) is planned. However, there is the possibility that the existing "changed" lines can be accidently deleted from objects. When users want to update an object, they do a whois look-up and re-direct the output into a temporary file. Then, they edit the file and send it to <auto-dbm(a)ripe.net>. Thus, before the output of the whois server was modified, the update contained all the previous changed fields. Now, unless the user does "whois -c", the "changed" attributes are not shown. But when users want to update an object, they re-direct a whois look-up into a file, edit it and then send it to the database. However, this time, the "changed" lines are missing and the updated version of the object does not contain the previous "changed" lines. This behavior is what would have happened previously if you had sent an update to the db and had deliberately omitted the changed lines. (BTW, if you send an update to the db which only has a new "changed" attribute or has all the old "changed" attributes replaced with a new one, a NOOP is the result. The object in the db is not changed). If all the original "changed" information in an inetnum object is not retained, there may be a delay in getting reverse delegation for the associated address space. This is because <auto-inaddr(a)ripe.net> checks the date when the assignment was originally made and compares it to the assignment window which the local registry had on that date. Thus, the request for reverse delegation is processed manually. So, to keep all the "changed" attributes, do 'whois -c' when you are making a copy of the object before updating it. If you have any questions, please contact <ripe-dbm(a)ripe.net>. Regards, Ambrose Magee ____________________________ RIPE Database Administration.

2 1

Re: Cc: in <auto-dbm> replies
by kissg＠sztaki.hu 11 Sep '98

11 Sep '98

> > I think the database should send any outgoing mails > > with sender address <nobody(a)no.where.ripe.net> or similar Dear Carsten, > come on, you're kidding, aren't you? This would mean to expand > Usenet's bad habits to RIPE email answers. That doesn't seem to > be a really good idea... I just say that *automatic* aswers should be blocked. However I don't think to hide the real sender as spammers do. Any human intelligence can find out easily where the mail came from if the RFC822 header contains a Sender: or X-This-Mail-Really-Came-From: ;-) line. So this header looks absolute correct for me: To: <as required> Cc: <as proposed> From: bit-bucket(a)ripe.net (Do not reply this address) Sender: ripe-db(a)ripe.net (RIPE database engine) X-Comment: This is an acknowledgement, not intended to be replied. Regards Gabor

1 0

Re: Cc: in <auto-dbm> replies
by kissg＠sztaki.hu 11 Sep '98

11 Sep '98

> The suggestion and motivation is excellent, but there are nasty side > effects in practice: > > Including the Cc: field is a nice way to create mailloops. > When the 'Cc:' address would be used too, there is no way for the > database to break the loop since the 'Cc:' address is set by the user > and can possibly contain the automatic mail address of the db and/or > a (local) alias to the db. Hi, David! I think the database should send any outgoing mails with sender address <nobody(a)no.where.ripe.net> or similar nonexistent address. This would break any mailing loop. This is acceptable for humanoid recipients too because generally they don't want to reply the notifications and acknowledgements. Regards Gabor

3 2