[Apologies for duplicate e-mails] Dear colleagues, This morning, between 07:00 and 09:00 (UTC), we deployed new software for the RIPE Database. Three changes were made: 1. The new software allows the creation of 32-bit (4-byte) Autonomous System Numbers (ASNs). There is more information about this in the references at the end of this e-mail. 2. The ORGANISATION object type NON-REGISTRY has been changed to OTHER. The RIPE NCC modified all existing organisation objects which had the old type. There is more information about this at: http://www.ripe.net/ripe/maillists/archives/db-wg/2006/msg00245.html 3. We reversed changes made to the status of ERX objects. While implementing the ASSIGNED ANYCAST status, other changes were made. This prevented more specific assignments from being created within EARLY REGISTRATION ranges. These changes have been rolled back. If you have any questions about any of these changes, send an e-mail to <ripe-dbm@ripe.net>. Regards Denis Walker Database Group RIPE NCC -------- References: The new 32-bit ASN format is described in these documents: http://www.ietf.org/internet-drafts/draft-michaelson-4byte-as-representation... http://www.ietf.org/internet-drafts/draft-uijterwaal-rpsl-4byteas-ext-01.txt Sites using the RIPE Database software will probably have to upgrade their systems. See: http://www.ripe.net/news/asn-db-update.html Our test database (e-mail to <test-dbm@ripe.net> or access through webupdates) also now handles the new AS number format.
On Jan 02, Denis Walker <denis@ripe.net> wrote:
This morning, between 07:00 and 09:00 (UTC), we deployed new software for the RIPE Database. Three changes were made: What happened to the planned changes about when irt objects are returned?
-- ciao, Marco
Hi Marco, Marco d'Itri wrote:
On Jan 02, Denis Walker <denis@ripe.net> wrote:
This morning, between 07:00 and 09:00 (UTC), we deployed new software for the RIPE Database. Three changes were made:
What happened to the planned changes about when irt objects are returned?
it is - as a 1st step - already being returned by default for web-queries. Just go to www.ripe.net and do a database lookup for e.g. 131.130/16 in the Database Search box, top right corner. Btw, to everyone just reading along :-) This lookup interface is pretty cool as it does support including your favourity whois flags like -T, -r and friends in the lookup key entry-box! Wilfried.
Wilfried Woeber, UniVie/ACOnet wrote:
Marco d'Itri wrote:
On Jan 02, Denis Walker <denis@ripe.net> wrote:
This morning, between 07:00 and 09:00 (UTC), we deployed new software for the RIPE Database. Three changes were made: What happened to the planned changes about when irt objects are returned?
it is - as a 1st step - already being returned by default for web-queries.
Which just does a local "whois -- <quote-escaped options>" and formats it in a pretty way.
Just go to www.ripe.net and do a database lookup for e.g. 131.130/16 in the Database Search box, top right corner.
That trick works at the moment only when the object has a direct IRT object related to it. Take for instance 2001:7b8:300::/48 and it won't return it, unless you add the -c option. Which is prolly why you said '1st step' as that will be the 2nd step. But the 1st step has already been in place for quite some time afaik...
Btw, to everyone just reading along :-) This lookup interface is pretty cool as it does support including your favourity whois flags like -T, -r and friends in the lookup key entry-box!
IMHO the cool part is the "Advanced" box so that one can select the flags and construct a query, allowing me to forget all those options ;) Greets, Jeroen
Hi Jeroen! Jeroen Massar wrote: [...]
That trick works at the moment only when the object has a direct IRT object related to it.
In the v4-world it works as expected, e.g. 193.171.3.0 - 193.171.3.255 does NOT have a direct link to our irt object, but it gets returned as expected (by me :-) from the /16 (or /15?))
Take for instance 2001:7b8:300::/48 and it won't return it,
As the /40 has a link to the irt: I consider that behaviour a bug ;-)
unless you add the -c option. Which is prolly why you said '1st step' as that will be the 2nd step. But the 1st step has already been in place for quite some time afaik...
Actually the 2nd step was discussed as changing the default behaviour of the commandline interface, too. The general feeling was that most queries by the walk-in customers would be run through the web interface anyway, and the NCC was reluctant to immediately change the interface which is probably used by scripts and programs...
Btw, to everyone just reading along :-) This lookup interface is pretty cool as it does support including your favourity whois flags like -T, -r and friends in the lookup key entry-box!
IMHO the cool part is the "Advanced" box so that one can select the flags and construct a query, allowing me to forget all those options ;)
Greets, Jeroen
Wilfried
On Jan 03, "Wilfried Woeber, UniVie/ACOnet" <Woeber@CC.UniVie.ac.at> wrote:
Actually the 2nd step was discussed as changing the default behaviour of the commandline interface, too. The general feeling was that most queries by the walk-in customers would be run through the web interface anyway, and the NCC was reluctant to immediately change the interface which is probably used by scripts and programs... This was discussed only at a WG meeting and reversed the mailing list consensus which had been established a few weeks before. Since then I have asked NCC to provide statistical data instead of feelings but I have not seen any yet.
Again: NCC, can you please publish some statistics about distribution of clients strings? Pretty please? I would like to see separate tables for all queries and for "naked IP with no flags (except -V)" (which is the case we care about). Also, I still believe that the whois page should have a huge link to a tool like http://www.cert.pl/cgi-bin/ipdig.pl, which uses formal attributes and then some euristics to find the abuse contact for an IP (and does not display anything else). -- ciao, Marco
Jeroen Massar wrote:
Wilfried Woeber, UniVie/ACOnet wrote:
Marco d'Itri wrote:
On Jan 02, Denis Walker <denis@ripe.net> wrote:
This morning, between 07:00 and 09:00 (UTC), we deployed new software for the RIPE Database. Three changes were made:
What happened to the planned changes about when irt objects are returned?
it is - as a 1st step - already being returned by default for web-queries.
Which just does a local "whois -- <quote-escaped options>" and formats it in a pretty way.
Just go to www.ripe.net and do a database lookup for e.g. 131.130/16 in the Database Search box, top right corner.
That trick works at the moment only when the object has a direct IRT object related to it. Take for instance 2001:7b8:300::/48 and it won't return it, unless you add the -c option. Which is prolly why you said '1st step' as that will be the 2nd step. But the 1st step has already been in place for quite some time afaik...
Yes the first step has been implemented some time ago. But I think you have highlighted a bug here with irt in inet6num objects. The web interface is still not finding them without adding the -c flag. We will look into this. regards Denis Walker Database Group RIPE NCC
Btw, to everyone just reading along :-) This lookup interface is pretty cool as it does support including your favourity whois flags like -T, -r and friends in the lookup key entry-box!
IMHO the cool part is the "Advanced" box so that one can select the flags and construct a query, allowing me to forget all those options ;)
Greets, Jeroen
On 2 Jan 2007, at 15:51, Denis Walker wrote:
2. The ORGANISATION object type NON-REGISTRY has been changed to OTHER. The RIPE NCC modified all existing organisation objects which had the old type. There is more information about this at:
http://www.ripe.net/ripe/maillists/archives/db-wg/2006/msg00245.html
Several of us in the ENUM WG are very glad to see this. Thanks and Happy New Year to everyone involved in making it happen. Best regards, Niall O'Reilly Co-Chair, RIPE ENUM Working Group
participants (5)
-
Denis Walker
-
Jeroen Massar
-
md@Linux.IT
-
Niall O'Reilly
-
Wilfried Woeber, UniVie/ACOnet