- db-wg - mailman.ripe.net

1. Draft, proposed agenda DB-WG meeting RIPE-34, Thu.,Sept.23'99
by Wilfried Woeber, UniVie/ACOnet 16 Sep '99

16 Sep '99

Attached please find the 1. draft for a proposed DB-WG agenda! As usual, additional input and comments are solicited! Please be prepared to see an update to that being circulated on the DB-WG list between now and just before the meeting. Looking forward to see many of you in Amsterdam, next week, Wilfried. Note to <Meeting(a)RIPE.NET> and <Webmaster(a)RIPE.NET>: Please use this version for the hand-outs and for the meeting web page. Thank you! -------------------------------------------------------------------------- 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 __________________________________________________________________________ Proposed Agenda for the Database Working Group Meeting RIPE-33, September 1999, Amsterdam, NL 1. Draft [ Note: for any updates between now and before the WG meeting, watch the DB WG Mailing List, please! ] A. Administrative stuff (Wilfried Woeber, chair) B. RIPE DB, operational update (RIPE NCC) C. DB SW Re-implementation (RIPE NCC) D. DB Consistency Project - presentation ((RIPE NCC) - discussion E. Birth of, and Life after the release of The New SW - migration/implementation planning (if any :-) - next development steps W. Do we need/want a DB WG review/re-structuring? - future topics of interest (workplan/charter?) - I/O with other WGs - logistics X. Report(s) Y. Input from other WGs - t.b.d. Z. AOB ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ WW, 16-SEP-1999 13:17:19

1 0

Re: RIPE handles delay
by Wilfried Woeber, UniVie/ACOnet 15 Sep '99

15 Sep '99

Dear Vesa! => response to updates more directly related to the RIPE DB intended purpose => while still ensuring that others not so directly related will get => processed. = =This tactful comment probably means a turn to the right direction. =But I wish it was said loudly and clearly, maybe even as a complete =change in DB usage policy. We pay for address registration services, =please take the domain (non-rev-dns-related) objects elsewhere. To statr with I'd like to assure you that we are (sometimes painfully) aware of the issue(s) you'r point fingers at. Quite a bit of things have been achived already to allow us to move in the right direction in due time: Technically: the NCC already has implemented the WG-approved refer: attribute to allow referrals to "remote" autoritative servers for domain: objects. This is already being use by (a few) TLD registries, others are working towards that end. Logistics: CENTR was formed within the framework of the RIPE NCC quite a while ago to deal with TLD activites (or DNS issues in general). CENTR has meanwhile been incoprorated as a legal entity of it's own, and moved office to the UK. RIPE discussions: we have already (recently) started to discuss (both informally amongst the WG chairs, as well as within the various WG meetings) the issues with co-locating different registries within a single database environment (i.e. the IP-Registry, Routing-Registry, DNS-Registry). These issues are (amongst others) . authority for data and responsibility for data quality . access patterns ie.e. (updates vs. consumers, privacy, copyright) . providing operational resources As an aside, you are correct in stating that the NCC's main objective rests with the IP-Address, including reverseDNS and Routing Registry activites. But, at the same time, the creation of and support for CENTR and the DNS-related activites was agreed by the RIPE community and there was (still is?) a contract in place between the NCC and CENTR to have CENTR contribute equitably to the NCC's operational costs. I'm looking forward to continue this thread of discussion (within the DB-WG framework or somewhere else) as well as to corrections and amendments by the RIPE community and WG chairs, and/or from NCC staff. Best regards, and thanks for bringing this up (again) for consideration! Wilfried Woeber DataBase WG chair. -------------------------------------------------------------------------- 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: RIPE handles delay
by Joao Luis Silva Damas 14 Sep '99

14 Sep '99

Hi, we have been having a lot of trouble lately here at the NCC. Apart from some HW problems, that are now fixed, we are still receiving updates at a rate and in a combination that the DB can't presently handle in a fast way. All messages sent are queued and processed in turn. We are working on optimizing the software to handle these updates faster. Also we are going to deploy a queue control system to ensure a prompt response to updates more directly related to the RIPE DB intended purpose while still ensuring that others not so directly related will get processed. Currently we are processing the mails that were causing the problems with a lower priority to ensure an adequate service to the community. I will be presenting an in depth report on the situation next week at the RIPE meeting (DB wg). Shall the TLD wg also be interested in the some sort of presentation or overview I will be glad to present it. We have been reporting on the situation to this and other RIPE wgs and shall continue to do so. At this moment the situation is such that messages are being processed as in the good old days :-) Best regards, Joao Damas RIPE DB Group Manager RIPE NCC On Tue, 14 Sep 1999, Alfredo E. Cotroneo wrote: > Hi, > > For the last ten days or so we have noticed that replies for a new contact > handle to auto-dbm(a)ripe.net is taking much more than the usual 2-5 minutes > as it used to be in the good old times ;-) Although we request only a few > handles per month, a relly from RIPE takes now 10-20 hrs or more in some > cases, regardless of the time (day or night, week or week-end) when the > request is sent out. > > I have heard of some kind of explanation related to SW/HW upgrades at RIPE > and huge requests of handles being queued by some provider a few days ago, > however, can someone at RIPE tell if the situation is going to be solved > soon and when ? > > Unfortunately this is causing at least an additional 1+ day delay in the > registration of new domains at the Italian NIC, as IT-NIC requests that a > handle exists for each contact before the domain registration form is > submitted. Of course any comment from IT-NIC on how they can modify their > internal procedures to minimize such delays (i.e. by IT-NIC requesting a > new handle to RIPE directly) is also welcome. > > Thanks. > > > -- > Alfredo E. Cotroneo PO Box 11028, 20110, Milano, Italy > Phone: +39-335-214-614 (try first) / +39-02-266-6971 > email: alfredo(a)nexus.org fax: +39-02-706-38151 > >

2 1

hide the dates on Ripe Database
by roberto.bazan＠inycom.es 09 Sep '99

09 Sep '99

Hi, everybody, I'am new in the list , I would like to know, if can i hide Ripe Database's date ? Thanks in advance

1 0

RIPE database
by Jerzy.Pawlus＠cyf-kr.edu.pl 07 Sep '99

07 Sep '99

[ Moderator note: changed local-ir(a)ripe.net -> db-wg(a)ripe.net ] Anybody knows what is happening with RIPE database? I have sent my update 24 hours ago and they are not visible. Jerzy Pawlus

1 0

Top 100 Maintainers List
by RIPE Database Administration 06 Sep '99

06 Sep '99

3 2

Slow performance of Ripe DB ?
by Paul Rolland 06 Sep '99

06 Sep '99

Hello, It seems we are currently facing a performance problem on the RIPE Db, which started at least yesterday... I sent an update request dated : Date: Sun, 5 Sep 1999 15:43:46 +0200 and got no answer since... Since the updates requested are not processed, I guess that either the mail has not been received or the processing delays are getting quite high... Maybe this is to be related to the fact that the maintainer report is still not published... On a general basis, I have the feeling that since this report is being generated, response time of ripe db has been seriously impacted... Is this just a bad feeling ? Regards, Paul Paul Rolland, rol(a)oleane.net France Telecom Oleane/Direction Technique/Directeur France Telecom Oleane/Technical Direction/Director -- Please no MIME, I don't read it - Pas de MIME, je ne le lis pas Please no HTML, I'm not a navigator - Pas d'HTML, je ne suis pas un navigateur "We all want to change the world" - The Beatles - Revolution

2 1

Delay in processing updates
by Engin Gunduz 06 Sep '99

06 Sep '99

Dear Colleagues, Currently we have a long mail queue to be processed by the RIPE database. If you haven't received an acknowledgement for your updates, there is no need to send them again, all messages to auto-dbm have been queued. We are sorry for the inconvenience, Regards, Engin Gunduz ____________________________ RIPE Database Administration.

1 0

Problems with this morning's results ?
by Paul Rolland 03 Sep '99

03 Sep '99

Hello, It seems that there is a problem with the results currently displayed on the Web in the Per Maintainer Inconsistency Report. OLEANE-NOC is mentionned to have 2908 NEVER_REFD, but it seems this is due to the fact that all the objects referenced in a .fr domains have been considered as NEVER_REFD, even though the domain seems to be in the RIPE DB. For example, we have Christophe Wagnon that is never ref'd, but it is used in the apore.fr domain. Regards, Paul Paul Rolland, rol(a)oleane.net France Telecom Oleane/Direction Technique/Directeur France Telecom Oleane/Technical Direction/Director -- Please no MIME, I don't read it - Pas de MIME, je ne le lis pas Please no HTML, I'm not a navigator - Pas d'HTML, je ne suis pas un navigateur "If you're not scared, you don't understand" - Mike O'Dell, VP UUNet Tech. Inc.

2 1

Sad news
by Joao Luis Silva Damas 27 Aug '99

27 Aug '99

Dear all, It is with great heart pain that I bring you some very sad news. Last Tuesday, our friend and colleague Roman Karpiuk suffered a terrible traffic accident while on holidays in his home country that has caused his death and that of his wife, Danka. The whole of the RIPE NCC is shocked after this event. I think I speak for the whole RIPE community if I say that we'll miss Roman and remember him as the joyful, creative and always ready to help colleague that he was. We shall always remember both of them with fond memories. Yours, Joao Luis Silva Damas RIPE NCC DB Group

1 0

Delay in processing updates
by RIPE Database Administration 25 Aug '99

25 Aug '99

Dear Colleagues, Due to an unprecedented series of major operational problems over the last few days, there are now many updates waiting to be processed by the RIPE Database. The waiting time for some updates has reached 24 hours. We have worked to fix this problem and the situation is improving. We expect the performance of the RIPE Database to be back to normal late tonight. In the meantime, we would ask you to be patient. If you have not yet received an acknowledgement to your updates, there is no need to send them again. All updates received by auto-dbm(a)ripe.net have been queued. We realise that is causing some inconvenience to you, but we would appreciate your co-operation. If you have any questions, please contact <ripe-dbm(a)ripe.net>. Regards, Marek Bukowy ____________________________ RIPE Database Administration.

1 0

Top 100 Maintainers List - 19990820
by RIPE Database Manager 24 Aug '99

24 Aug '99

4 5

Top 100 Maintainers List
by RIPE Database Manager 04 Aug '99

04 Aug '99

2 1

Re: Interesting object in the RIPE database
by Steven Bakker 11 Jul '99

11 Jul '99

>>>>> On Wed, 7 Jul 1999, "RLE" == Robert_Martin-Leg=E8ne wrote: RLE> Doesn't mean that it makes sense in the IP Registry world. Let's be RLE> practical. And besides, classful addressing is a Thing of the Past (ahem). The BSD way of doing it is clever, but complicated and it has no place in the DB, IMO. I always understood that the DB software would pad missing octets at the end of an IP address, no ambiguity there. What I'm surprised about is that the DB apparently allows someone to register address space that's most likely not even theirs... Of course, once it gets allocated and entered into the DB, a conflict will be detected and flagged (right!?), but until then, the error will go undetected. Is this a feature? Cheers, Steven

4 3

Interesting object in the RIPE database
by Niels Bakker 07 Jul '99

07 Jul '99

Hi, I noticed the following object in the RIPE database this afternoon: inetnum: 195.59.213.0 - 213.127.0.255 netname: WILLIAMBROOKES-TF13-GW descr: William Brookes School country: GB admin-c: PC2598-RIPE tech-c: IH22-ORG rev-srv: ns0.insnet.net rev-srv: ns1.insnet.net status: ASSIGNED PA mnt-by: AS5378-MNT changed: cliveg(a)insnet.net 19990706 source: RIPE (I stumbled over it while querying the RIPE database for an object within the APNIC database while accidentally omitting the -a option.) Clive Galway (the creator of this object) told me after I contacted him that he had entered the following information into the template: 195.59.213.0 - 213.127 instead of 195.59.213.0 - 195.59.213.127 He added, "funny that it dont complain though" - I agree wholeheartedly with him here. :) A check for the number of dots in the inetnum: line should be quite easy, and similar to this: # fix for people who don't include the first two or three octets s/^(\d+\.\d+)\.(\d+\.\d+) - (\d+\.\d+)$/$1.$2 - $1.$3/ s/^(\d+\.\d+\.\d+\).(\d+) - (\d+)$/$1.$2 - $1.$3/ if ($_ =~ /^\d+\.\d+\.\d+\.\d+ - \d+\.\d+\.\d+\.\d+$/) { &complain; } I believe there is a sub to check whether something is a valid IP address but can't find it offhand in my (by now pretty dated) copy of the RIPE database software... With regards, -- Niels Bakker, * * EuroNet Internet BV Network Operations * * Herengracht 208-214 * 1016 BS Amsterdam NJB9 * +31 (0)20 535 5555

3 2

Converting named references to NIC handle references
by Engin Gunduz 29 Jun '99

29 Jun '99

Dear collegues, As discussed and agreed to in the mailing lists, we will be converting unambiguous named references in the admin-c and tech-c fields of as-macro, autnum, community, dom-prefix, inet-rtr, mntner and role objects to NIC handle references, on 1 July, 1999, at 17:00, GMT. The process will take around half an hour, and during the process the updates will be queued. Whois queries will not be affected. The maintainers of the affected objects will get notifications about the changes. The conversion on domain and inetnum objects will be done at a later date, due to substantial time they require for this process. If you have any questions, please do not hesitate to contact us. Regards, Engin Gunduz RIPE NCC DB Group

1 0

T1
by OrderOffice - ICP Lueck und Lanzrath GbR 25 Jun '99

25 Jun '99

T1

1 0

address field limitation???
by Pedro Goncalves 25 Jun '99

25 Jun '99

Hi I would like to know what is the char limit (if any) for the address field. Some of my submitions apparently where returned because of the usage of more than 64 chars. Best regards Com os melhores cumprimentos. Pedro Gonçalves Telepac, SA Tel: 351-1-7907354 Fax: 351-1-7907001

2 1

Re: Modifying references to contact info in the RIPE DB
by Wilfried Woeber, UniVie/ACOnet 24 Jun '99

24 Jun '99

Hi Joao! >If the reference is by name then when I query the DB I will get the list of >persons with that name, right? Right. >Whether the object belongs to the person or not is not going to save this >person from being contacted. From the point of view of the database and the >user querying it, the object belongs to that person. Right >The link is via a name. > >Now we substitute the reference by name with a reference by nic handle >pointing to the same person. >What is the difference? > >ALMOST NONE except that: True, for the case where there is a strict one-to-one mapping, e.g the recursion using the name returns *one* object. What do you suggets to do with the case where the recursion, based on the names, returns a *set* of objects? I can see 2 approaches (at least): - do not perform the handle substitution for that referring object and put it on the list of those requiring human intervention - expand the xxxx-c: list (which? the tech-c:?) to refer to *all* objects, but by ref'ing their handles While human beings are quite clever in selecting the "proper" instance of a person: object (by looking at country, address, email,...) a script might be quite clueless with that respect :-) 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 --------------------------------------------------------------------------

3 2

Re: Modifying references to contact info in the RIPE DB
by Wilfried Woeber, UniVie/ACOnet 24 Jun '99

24 Jun '99

>I explicitly excluded those objects from this run. Great. >I knew this thing was going to be a bit polemic so I am trying to split into >smaller problems and stages that we can all agree to. Well, complex, at least. I fully support your approach! >Hopefully every step will reduce the number of objects that need repairs. And >believe me, it's not the same to have to fix 250000 objects than to fix 25000. >Somehow the latter seems a bit more feasible (from a "let's deal with this" >attitude). Certainly. >So, can we agree to proceed with our suggestion as stated below? >** >Change references by names to references by nic handle where these are unique >(there are not already more than two persons with the same referenced name). >** I agree. >For the cases where the ambiguity is already there: we'll look at them, try to >categorize them Sounds interesting. I really like the approach taken by Poul-Henning with his IP address page. Maybe we can come up with a similar list, e.g. categorized by object type, and within the object type by LIR, TLD, AS-#, Maintainer... >and maybe identify another subset which can be handled by >drawing information from some other source. >We'll get back to you with our suggestion (and we take the two you mention as >input). Any input is most welcome. Wilfried.

1 0

Re: Modifying references to contact info in the RIPE DB
by Janos Zsako 24 Jun '99

24 Jun '99

> From owner-db-wg(a)ripe.net Thu Jun 24 15:23:02 1999 > From: Joao Luis Silva Damas <joao(a)ripe.net> > ALMOST NONE except that: > from that point on if someone else with a name equal to one already in the > database is entered into the database the reference will not be made ambiguous > while if we leave it as a reference by name it sure becomes ambiguous. I agree with Joao that there is virtually nothing to lose. I don't think it is better to have a reference by name with a single incorrect instance of a person object with that name, rather than a NIC-handle reference to the same incorrect instance... (By incorrect instance I mean a person object with the same name, but who has nothing to do with the referring object). At the same time the benefit is obvious (see above). Also, the longer one delays this process, the better the chances to have several instances to be handled manually (i.e. to try to identify the correct person object from several ones with the same name). Just my 2 filler (.01 HUF). :)) Janos

1 0

Re: Modifying references to contact info in the RIPE DB
by Janos Zsako 24 Jun '99

24 Jun '99

> From owner-db-wg(a)ripe.net Thu Jun 24 14:35:25 1999 > From: "Wilfried Woeber, UniVie/ACOnet" <woeber(a)cc.univie.ac.at> > One of the key things might be to make sure that we (you :-) can perform > a roll-back, if it turns out that we've messed up things badly... My feeling is that if one replaces the name reference with the NIC-handle, at least as long as the referenced person object is not deleted (which I suppose cannot be deleted once it has an other object referencing it), the roll-back is always possible (i.e. one could re-substitute the NIC-handle with the name). Although, I feel this would be in the best case useless, since either there is no other person object with the same name, or the other object(s) was(were) added later. Janos

1 0

Re: Modifying references to contact info in the RIPE DB
by Wilfried Woeber, UniVie/ACOnet 24 Jun '99

24 Jun '99

Hi Joao! >So, in spite of the concerns raised (valid as they are) I would still >vote to go ahead on this one as proposed. > >What do people think? Well, some as yet unstructured rants, off the top of my head... I'd love to see some sort of test-run being performed, e.g. going ahead and putting the result up as the (or an additional) test db? One of the key things might be to make sure that we (you :-) can perform a roll-back, if it turns out that we've messed up things badly... Is there any chance to do some sort of dry-run, and put up some sort of list of cases deserving investigation? Poul's page is a very good thing, indeed, although tackling a different aspect. My reasoning is that whatever can be fixed by humans, in a distributed project, before the brute-force attack would make life easier. Most probably basic rule #1 applies to my comments: use the brain before using the keyboard :-) 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 --------------------------------------------------------------------------

2 1

Modifying references to contact info in the RIPE DB
by Joao Luis Silva Damas 24 Jun '99

24 Jun '99

Dear all, the Database consistency project has been going forward in its goals since RIPE 33 (eg. addition of nic handles to all objects without one.) Statistics and reports about the "status" of the database can be seen at http://www.ripe.net/db/state/. We will keep adding information to these pages as the project progresses. We are now at the point where we would like to tackle the issue of references by name in the admin-c, tech-c and zone-c. As a next step, we would like to have references by nic handles (which are unique) instead of references by name (which can be non -unique). At this point, we would like to automatically go through all the references by name which can uniquely be assigned to a person or role object and substitute them by references to their nic handles. This will prevent these references from becoming non-unqiue in the future. Later we will go through the process of fixing the non-unique references left in the database. This process requires user intervention since we can't decide who "owns" a particular set of resources. For this later stage, we will be preparing reports that we will send to all maintainers for whom we detect objects with inconsistencies and try to get them fixed. Information about this part will be sent to the lists in a few weeks. We would like to do this after as much as possible has been done by automatic means to correct inconsistencies so that human effort is reduced to a minimum. If there are any concerns regarding this move, please make them known before next Friday (June 25th). Looking forward to your comments, Joao Damas Database Group RIPE NCC

4 6

Changing old format dates in the RIPE DB
by Joao Luis Silva Damas 24 Jun '99

24 Jun '99

Dear all, we are going to start with the procedure to modify all the dates in the changed fields of database objects that were using only two figures for the year. After the modifications, the year part of the date will contain 4 digits. Changed fields which currently do not have a date will not be changed (*) This is taking place according to the action points agreed on the Vienna RIPE meeting. The process will be carried over the next few days due to the time it takes to re-index some of the bigger files. Regards, Joao Damas DB Group RIPE NCC (*) These objects will not be updatable by the user unless a date is included at update time or the only changed field without a date is the last one, in which case the database automatically fills in the date at which the update takes place.

1 0