December 2022 - db-wg - mailman.ripe.net

Using the RIPE Database for IPAM
by denis walker 04 Oct '23

04 Oct '23

Colleagues Do you use the RIPE Database as your IPAM solution? If so, read on.... The RIPE NCC has done a brief review of the recommendations made by the DB Task Force (DBTF). You can find it here: https://www.ripe.net/participate/ripe/tf/rdb-requirements-tf/initial-analys… There is one point in the summary that I disagree with: One recommendation can be implemented immediately by the RIPE NCC: - 3. Using the RIPE Database as an IPAM solution The recommendation from the DBTF was: "The task force recommends limiting and discouraging the use of the RIPE Database as an enterprise IPAM solution." This has not yet been discussed by the community and there is no consensus on this recommendation. So the RIPE NCC should not be doing anything to implement this recommendation. There are a number of points relating to this. The first BoF held by the DBTF had 46 participants. They had some polls about IPAM: Where do you hold your canonical IP address assignment details? (13 answers) Internal database or IPAM system (5 - 38%) RIPE Database (5 - 38%) Internal database for detailed assignments and RIPE database for allocations (3 - 23%) What should we do about the IPAM use of the RIPE Database? (12 answers) We should discourage it (3-25%) We should tolerate it but provide restrictions and best practices (5-42%) We should promote it and facilitate it (4-33%) During the BoF some comments were also made on this: Shane's comment "IPAM seems to be an important thing" Nathalie said about 50% of people attending training courses over a number of years said they use the RIPE DB for IPAM. Also trainees said open source IPAM solutions often pull data out of the RIPE DB but have no functionality to push data to the RIPE DB. During the presentation by the DBTF in the DB-WG session at RIPE 83 Shane said (on IPAM): "if you had very light requirements for IPAM then maybe it is OK, but it's not really what it's for" "unable to find out if this [response] was based on misunderstanding of the question" "we think there are probably good ways to go forward with that, for example making hooks so that external IPAM solutions can hook into it...that makes sense, there are certainly many commercial vendors that make IPAM products who would be happy to use hooks, there are many ways that third parties could use this but probably not a requirement" So it looks like there are very mixed views on the use of the RIPE Database for IPAM. Some people also expressed surprise that so many people use the database for IPAM. It is clear that we don't actually understand who, how and why people use the database for IPAM. We need a better understanding of the how and why and the number of users doing this before we can make any decision about discouraging or encouraging this use of the database. Even to consider if IPAM should be defined as a purpose of the database. As we are talking about it, maybe now is the time to start this discussion. If you use the RIPE Database as your IPAM solution maybe you can offer some more details. cheers denis co-chair DB-WG

5 5

NWI-14 Authenticating References to Objects
by denis walker 19 Jun '23

19 Jun '23

Colleagues The chairs propose creating NWI-14 for expanding the optional use of the "mnt-ref:" attribute to protect objects from accidental or intentional mis-representation of responsibility. See slide 14 of this presentation for the problem definition: https://ripe78.ripe.net/presentations/129-RIPE78-Operational-Update.pdf We ask the RIPE NCC to add NWI-14 to the list. cheers denis co-chair DB-WG

3 4

Proposed Service Criticality Form
by Edward Shryane 12 May '23

12 May '23

Dear Colleagues, During RIPE 84 in May, we discussed the Service Criticality Framework. We mentioned that we would like the Database Working Group's input on the criticality rating for the RIPE Database. Thank you to the Co-chairs for asking the working group for input in July with the original form: https://www.ripe.net/ripe/mail/archives/db-wg/2022-July/007527.html Since we have yet to receive any feedback so far, we plan to share our proposed completed form with the working group on Monday, 12 December. We hope that this will invite more input. Regards Ed Shryane RIPE NCC

3 4

RIPE-NONAUTH AS-SET bug fix
by denis walker 20 Feb '23

20 Feb '23

Colleagues There is some support for the idea that if AUT-NUM objects in RIPE-NONAUTH authorise the creation of AS-SET objects, these set objects will also be in RIPE-NONAUTH. There is also support for this to be considered as a bug from the implementation of 'NWI-5 Out of region ROUTE(6)/AUT-NUM objects'. So existing AS-SET objects whose creation was authorised by one of these RIPE-NONAUTH ASNs can be moved to RIPE-NONAUTH as part of a bug fix. Does anyone have any objections to such a 'bug fix'? cheers denis co-chair DB-WG On Thu, 1 Dec 2022 at 19:34, Nick Hilliard <nick(a)foobar.org> wrote: > > Cynthia Revström wrote on 30/11/2022 22:59: > > I am not sure if this feature is used or not however I think this is a > > very good reason to not go forward with a clean-up (at least until we > > have properly evaluated things). > > We will probably have to figure out some other way to deal with > > objects that are currently causing issues I think. > > the "feature" is used, yes. Some providers have customers in different > RIR service regions. Some organisations have address space registered > in different RIR service regions. It's impossible to avoid in many > situations. > > What's important right now is to close off the option to create new > unqualified as-set names, and to move the existing qualified non-RIPE > ASxxxx:as-set objects from source: RIPE to source: RIPE-NONAUTH. > > Denis was correct that this was a bug during the implementation of NWI-5 > (not ripe-731 which I mistakenly quoted). > > After that, we can afford to spend a bit of time looking at potential > clean-up options. There are 1590 empty as-set objects. 700 of these > haven't been updated in the last 5 years, and some going back 20 years. > > I wouldn't lose too much sleep about deleting empty as-sets. Contact > people, set a timeout, and then delete. Worst case, people can > reference new, qualified as-sets. > > Nick >

4 8

Extending the consultation period for service criticality
by Razvan Oprea 18 Jan '23

18 Jan '23

Dear colleagues, Your input about the criticality rating of the RIPE Database is important. Keeping in mind the busy end of year, and the holidays, we have extended our consultation period until 22 January 2023. Wishing you all a happy new year. Regards, Razvan C. Oprea RIPE NCC

1 1

Impact Analysis for NWI-4
by Edward Shryane 16 Dec '22

16 Dec '22

Dear colleagues, Following is the impact analysis for NWI-4 "Role of status: field in multivalued status context", based on the addition of a new status value "ALLOCATED-ASSIGNED PA". The main drawback identified in this proposal is that an inetnum with status "ALLOCATED-ASSIGNED PA" must remain co-maintained between the RIPE NCC and the LIR, and not by the end user. If an end user is required then the earlier alternative proposal to use a tuple (using both prefix and status) is preferable (although technically more complex). Whois Application * Add a new inetnum status "ALLOCATED-ASSIGNED PA". Treat this status as both ALLOCATED PA and ASSIGNED PA. * New allocations will continue to be created with the "ALLOCATED PA" status. * Allow a user to switch the status to and from (only) "ALLOCATED PA" to "ALLOCATED-ASSIGNED PA" themselves. This allows the user to document a single assignment under the allocation. * Follow the same validation rules for both ALLOCATED PA and ASSIGNED PA. This new status has no impact on the validation rules (i.e. no conflict was found between the two status values). * Support the same status hierarchy for ALLOCATED-ASSIGNED PA as for ALLOCATED PA or ASSIGNED PA (e.g. SUB-ALLOCATED PA, LIR-PARTITIONED PA etc.). * An inetnum with status "ALLOCATED-ASSIGNED PA" remains co-maintained by the RIPE NCC. * The inetnum primary key remains the IPv4 address range. Using the same address range with different status values is not allowed. Registry Services * The "IPv4 Address Allocation and Assignment Policies for the RIPE NCC Service Region" policy (currently RIPE-733) requires that "All assignments and allocations must be registered in the RIPE Database". This proposal addresses this requirement. * If an LIR wants to create an assignment because they want to document an assignment to an end user, this proposal will not help in this case. The ALLOCATED-ASSIGNED PA inetnum remains co-maintained between the LIR and the RIPE NCC, and not by an end user maintainer. Also any contacts (admin-c, tech-c, abuse-c etc.) are for the LIR and not the end user. * What is missing is that an LIR can not specify an end user customer’s mnt-by and, most importantly, cannot specify the customer’s organisation object. Internal Registry * No impact is expected to the internal registry software. No direct parsing of the status attribute for either inetnum objects was found in interactions with Whois (e.g. organisation changes and transfers). Whois Clients * Clients reading the inetnum status: value must be updated to take account of the "ALLOCATED-ASSIGNED PA" status and treat it as equivalent to both an "ALLOCATED PA" and "ASSIGNED PA" status. Regards Ed Shryane RIPE NCC

1 0

RIPE Database Quarterly Planning Q1 2023
by Edward Shryane 15 Dec '22

15 Dec '22

Dear colleagues, We have published our quarterly plans for Q1 2023 for the RIPE Database: https://www.ripe.net/manage-ips-and-asns/db/quarterly-planning In this quarter, we will continue working on the server-side implementation of NWI-12: NRTMv4 and the migration of the RIPE Database documentation to a new interface. We hope that by providing these plans, we can better inform the community of the ongoing developments on the RIPE Database and get feedback on our work. If you have comments or questions around this, we hope you'll let us know on the list or you can always contact the DB team directly at ripe-dbm(a)ripe.net . Regards Ed Shryane RIPE NCC

1 0

Whois Release 1.105
by Edward Shryane 13 Dec '22

13 Dec '22

Dear Colleagues, RIPE Database release 1.105 has been deployed to the Release Candidate environment. We plan to deploy to production on Tuesday 13th December. This release includes a single change from 1.104: * NWI-19: Change to AS-SET object naming rules (not allowed to create with short format) The release notes are on the website: https://www.ripe.net/manage-ips-and-asns/db/release-notes/ripe-database-rel… More information on the Release Candidate environment can be found on our website: https://www.ripe.net/manage-ips-and-asns/db/release-notes/rc-release-candid… Please let us know if you find any issues with this release in the RC environment. Regards Ed Shryane RIPE NCC

1 1

NWI-10 "Definition of Country" is Complete
by Edward Shryane 02 Dec '22

02 Dec '22

Dear Colleagues, The RIPE NCC has now completed the implementation of phase 3 of NWI-10: "Definition of Country". This work included: * Recording the legal address for all End User organisations with Provider Independent resource(s); * Populating the legal address country code into end user organisation objects in the RIPE database; * Updating the country code in the Extended Delegated Stats for all the Provider Independent resources an end user holds to the legal address country code. * Internal software changes to synchronise changes of legal address country code to the RIPE database and Extended Delegated Stats. More information about NWI-10 is available here: https://labs.ripe.net/author/stefania_fokaeos/our-plan-to-update-country-co… Regards Ed Shryane RIPE NCC

1 0

proposal: disallow creation of new non-hierarchically named AS-SET objects
by Job Snijders 01 Dec '22

01 Dec '22

Dear DB-WG, Speaking in individual capacity. In RFC 2622 section 5 specifies the naming convention for AS-SET objects. https://www.rfc-editor.org/rfc/rfc2622#section-5.1 There basically are two styles: * "short" (example: AS-SNIJDERS) * "hierarchical" (example: AS15562:AS-SNIJDERS) Problem statement ================= In recent weeks a number of hypergiant cloud providers have faced the thorny effects of adversarial AS-SET object naming collisions between IRR databases. An example of this phenomenon is the existence of AS-AMAZON in both RADB and RIPE. According to https://www.peeringdb.com/net/1418 the RADB copy of the object is the the correct one and populated with a number of members entries. The RIPE one is empty, and not under control of Amazon. The existence of the AS-AMAZON object in the RIPE database might cause some operators to inadvertently apply empty prefix-filters to EBGP sessions which in turn causes various problems. It seems Amazon has no recourse to get the AS-AMAZON object removed from the RIPE database; because the existence of that object in the RIPE database does not violate any policies (as far as I know). But perhaps, going forward, this community can do a little bit more to help prevent similar situations from happening to others. Solution proposal ================= I think the solution is to - GOING FORWARD - disallow creation of new AS-SET objects which follow the 'short' naming style. The advantage of hierarchical naming is that the existing authorization rules as applied by the RIPE Whois Server database engine do a decent job of protecting/separating namespaces. 'Grandfathering' existing short-named objects ensures that implementation of this solution proposal causes minimal (if any) disruption to existing workflows. The RIPE database engine blocking creation of short-named AS-SETs might help nudge the industry towards making hierarchical naming the norm. Related work ============ Related work throughout the registry industry: IRRd version 4 forces new AS-SET objects to be structured hierarchically: https://github.com/irrdnet/irrd/issues/408 Kind regards, Job

21 50