Have you looked at: https://labs.ripe.net/Members/kranjbar/proposal-to-display-history-of-object... Not sure if this is still in Beta? I don't know the the process or policy of each RIR and am more familiar with ARIN. ARIN does not get rid of historical information. They keep the records and make the information available through a service called WhoWas. https://www.arin.net/resources/whowas/index.html --Heather On Thu, Feb 12, 2015 at 2:38 PM, William Sylvester < william.sylvester@addrex.net> wrote:

Hi,

I support this and also was wondering about associated objects like routing objects and in-addrs? This is also an issue for claiming an old block where a locked maintainer was used. It would be nice to have the unlocked maintainer apply for objects related to the block not just the inetnum for example.

Thanks, Billy

On Feb 12, 2015, at 2:33 PM, Erik Bais <ebais@a2b-internet.com> wrote:

Hi,

With the IP transfers going in full swing, it is quite common these days that the larger prefix size is broken up in to multiple smaller prefixes..

As the original prefix is split up, the IPRA’s remove the original prefix from the database .. and with it .. the history …

Is there a way to preserve that data to be able to still lookup the data using –list-versions or –show-version # …

Regards,

Erik Bais

Dear colleagues, the versions feature is no longer in beta. You can query for the version history of a resource, a previous version of the object, and the difference between any two versions of an object. However, this feature does not list the version history for a deleted object, or for any versions before an object was re-created. If you have any questions for the RIPE NCC, please contact us. Regards Ed Shryane RIPE NCC

On 13 Feb 2015, at 01:54, Heather Schiller <heather.skanks@gmail.com> wrote:

Have you looked at: https://labs.ripe.net/Members/kranjbar/proposal-to-display-history-of-object... <https://labs.ripe.net/Members/kranjbar/proposal-to-display-history-of-objects-in-ripe-database> Not sure if this is still in Beta?

I don't know the the process or policy of each RIR and am more familiar with ARIN. ARIN does not get rid of historical information. They keep the records and make the information available through a service called WhoWas. https://www.arin.net/resources/whowas/index.html <https://www.arin.net/resources/whowas/index.html>

--Heather

On Thu, Feb 12, 2015 at 2:38 PM, William Sylvester <william.sylvester@addrex.net <mailto:william.sylvester@addrex.net>> wrote: Hi,

I support this and also was wondering about associated objects like routing objects and in-addrs? This is also an issue for claiming an old block where a locked maintainer was used. It would be nice to have the unlocked maintainer apply for objects related to the block not just the inetnum for example.

Thanks, Billy

On Feb 12, 2015, at 2:33 PM, Erik Bais <ebais@a2b-internet.com <mailto:ebais@a2b-internet.com>> wrote:

...

Hi,

With the IP transfers going in full swing, it is quite common these days that the larger prefix size is broken up in to multiple smaller prefixes..

As the original prefix is split up, the IPRA’s remove the original prefix from the database .. and with it .. the history …

Is there a way to preserve that data to be able to still lookup the data using –list-versions or –show-version # …

Regards,

Erik Bais

Hello, is there any resolution about deleted object in history? Is there any reason to not publish these data? With regards, Daniel On 13.2.2015 16:54, Erik Bais wrote:

Hi Edward,

The deleted objects from the RIPE DB with history (--list-versions and --show-version #) is what I’m currently looking for.

For instance … ( I took the next particular inet-num for no specific reason but to specify a case where the object is no longer present in the database a the original inet-num .. )

93.112.0.0/13 Was as a prefix assigned to an LIR …

The LIR has been carving out prefixes to various other entities ..

But no history on the original prefix 93.112.0.0/13 can be found anymore …

% This object was deleted on 2014-03-18 15:01

If someone would like to know more about one of the new inet-num’s that came out of it .. Let’s say : 93.117.0.0/21 ..

whois -h whois.ripe.net -- "--list-versions 93.117.176.0/20"

rev# Date Op.

1 2014-08-22 15:45 ADD/UPD

That particular inet-num shows a creation date … but not where it came from .. and it also doesn’t link to the upper original /13 .. and the original /13 historic data is also gone.

In the case where someone might have a prefix that has been put on a blacklist for instance.. and that particular prefix has been on some blacklist for some years … split up in various other prefixes..

The new owner of the prefix can’t show that they are the original legitimate owner of the prefix .. and the blacklist operators that haven’t updated their RBL to the actual new reality will not move …

I’ve seen this with several cases with SORBS for instance .. and as there is no history anymore … it makes dealing with these companies increasingly difficult..

In the case we talk about PA space .. there is still some trail of information on the IPv4 Transfer page .. statistics .. however if we talk about Legacy space.. which is not listed there … you are stuck …

Regards,

Erik Bais

*Van:*db-wg [mailto:db-wg-bounces@ripe.net] *Namens *Edward Shryane *Verzonden:* vrijdag 13 februari 2015 11:34 *Aa**n:*db-wg@ripe.net *Onderwerp:* Re: [db-wg] History lost ...

Dear colleagues,

the versions feature is no longer in beta.

You can query for the version history of a resource, a previous version of the object, and the difference between any two versions of an object.

However, this feature does not list the version history for a deleted object, or for any versions before an object was re-created.

If you have any questions for the RIPE NCC, please contact us.

Regards

Ed Shryane

RIPE NCC

On 13 Feb 2015, at 01:54, Heather Schiller <heather.skanks@gmail.com <mailto:heather.skanks@gmail.com>> wrote:

Have you looked at: https://labs.ripe.net/Members/kranjbar/proposal-to-display-history-of-object... Not sure if this is still in Beta?

I don't know the the process or policy of each RIR and am more familiar with ARIN. ARIN does not get rid of historical information. They keep the records and make the information available through a service called WhoWas. https://www.arin.net/resources/whowas/index.html

--Heather

On Thu, Feb 12, 2015 at 2:38 PM, William Sylvester <william.sylvester@addrex.net <mailto:william.sylvester@addrex.net>> wrote:

Hi,

I support this and also was wondering about associated objects like routing objects and in-addrs? This is also an issue for claiming an old block where a locked maintainer was used. It would be nice to have the unlocked maintainer apply for objects related to the block not just the inetnum for example.

Thanks,

Billy

On Feb 12, 2015, at 2:33 PM, Erik Bais <ebais@a2b-internet.com <mailto:ebais@a2b-internet.com>> wrote:

Hi,

With the IP transfers going in full swing, it is quite common these days that the larger prefix size is broken up in to multiple smaller prefixes..

As the original prefix is split up, the IPRA’s remove the original prefix from the database .. and with it .. the history …

Is there a way to preserve that data to be able to still lookup the data using –list-versions or –show-version # …

Regards,

Erik Bais

Piotr Strzyzewski wrote:

On Mon, May 18, 2015 at 05:42:59PM +0200, Daniel Suchy wrote: [...] From my point of view, there was no formal proposal to discuss. Just an informal question by Eric. It is even unclear which objects are expected to be visible.

IIRC there were a couple of quite informal discussions about this Q, some of them maybe even in the hallways :-) The general line of reasonsing seesm to be: The Data Protection environment states that an organisation is not supposed to collect data, or to keep storing it (and even less so, making it publicly accessible!) unless there is a need to do so for the declared reason, operation or goal. As the obejcts have been deleted, it can be inferred that there is no longer a valid reason to store those items or to make them accessible.

Piotr

Nevertheless, I suppose the NCC would still be able to work with e.g. LAEs to disclose the transaction that removed an item from the DB ;-) FWIW, HTH, Wilfried

Den 2015-05-19 14:44, Piotr Strzyzewski skrev:

On Tue, May 19, 2015 at 12:45:12PM +0200, Wilfried Woeber wrote:

...

Piotr Strzyzewski wrote:

...

On Mon, May 18, 2015 at 05:42:59PM +0200, Daniel Suchy wrote: [...] From my point of view, there was no formal proposal to discuss. Just an informal question by Eric. It is even unclear which objects are expected to be visible. IIRC there were a couple of quite informal discussions about this Q, some of them maybe even in the hallways :-) I have to admit that I deliberately not mentioned them, since they were (as you wrote) informal. I do believe, that interested parties should move those informal discussions to this mailing list.

I distinctly remember that we were asked if we wanted to have the deleted objects viewable. In fact it was at RIPE 66 in Dublin. I'd like to think that this is formal enough to proceed to next level. Slide 14. https://ripe66.ripe.net/presentations/186-DB_Update-66.pdf 13m30s in https://ripe66.ripe.net/archives/video/1174/ I asked about it after RIPE 67. https://www.ripe.net/ripe/mail/archives/db-wg/2013-November/004193.html -- Bengt Gördén Resilans AB

I would love to be able to query 'deleted' objects similár as that we currently can with list-version and -show-version .. Data being deleted because a parent object is subnetted due to a transfer is not helping in some cases .. For instance, if you want to get ip ranges removed from $rbl .. They tend to have 5 years old data on their rbl list .. With the previous subnet size ... So in order to be able to proof that the new transferee is someone else than the person /company they listed ... It helps to have the old data .. Regards, Erik Bais Verstuurd vanaf mijn iPad

Op 20 mei 2015 om 18:58 heeft Piotr Strzyzewski <Piotr.Strzyzewski@polsl.pl> het volgende geschreven:

On Tue, May 19, 2015 at 05:19:55PM +0200, Bengt Gördén wrote:

Dear Bengt

...

Den 2015-05-19 14:44, Piotr Strzyzewski skrev:

...

On Tue, May 19, 2015 at 12:45:12PM +0200, Wilfried Woeber wrote:

...

Piotr Strzyzewski wrote:

...

...

On Mon, May 18, 2015 at 05:42:59PM +0200, Daniel Suchy wrote: [...] From my point of view, there was no formal proposal to discuss. Just an informal question by Eric. It is even unclear which objects are expected to be visible. IIRC there were a couple of quite informal discussions about this Q, some of them maybe even in the hallways :-) I have to admit that I deliberately not mentioned them, since they were (as you wrote) informal. I do believe, that interested parties should move those informal discussions to this mailing list.

I distinctly remember that we were asked if we wanted to have the deleted objects viewable. In fact it was at RIPE 66 in Dublin. I'd like to think that this is formal enough to proceed to next level.

Slide 14. https://ripe66.ripe.net/presentations/186-DB_Update-66.pdf

13m30s in https://ripe66.ripe.net/archives/video/1174/

I asked about it after RIPE 67. https://www.ripe.net/ripe/mail/archives/db-wg/2013-November/004193.html

And the answer was:

"There was no discussion or agreement on taking this any further at RIPE 67 or since on the mailing list. If the community would like us to provide this service, we can add this to our feature list for a future release of the RIPE Database software."

Since there was almost lack of discussion, my personal proposal is to first ask NCC legal department for legal comment about deleted objects being viewable. I would suggest limiting the list of possible objects to the same for which DB provides --show-version and --list-versions nowadays.

Is it ok with you and the rest of the WG?

Piotr

-- gucio -> Piotr Strzyżewski E-mail: Piotr.Strzyzewski@polsl.pl

On Wed, 20 May 2015, Bengt Gördén wrote: Agree.

Den 2015-05-20 19:36, Erik Bais - A2B Internet skrev:

...

I would love to be able to query 'deleted' objects similár as that we currently can with list-version and -show-version ..

Data being deleted because a parent object is subnetted due to a transfer is not helping in some cases .. For instance, if you want to get ip ranges removed from $rbl .. They tend to have 5 years old data on their rbl list .. With the previous subnet size ... So in order to be able to proof that the new transferee is someone else than the person /company they listed ... It helps to have the old data ..

+1

-- Mvh Fredrik Widell Resilans AB http://www.resilans.se/ mail: info@resilans.se , fredrik@resilans.se phone: +46 8 688 11 82

Den 2015-05-20 18:58, Piotr Strzyzewski skrev:

On Tue, May 19, 2015 at 05:19:55PM +0200, Bengt Gördén wrote:

Dear Bengt

...

...

On Tue, May 19, 2015 at 12:45:12PM +0200, Wilfried Woeber wrote:

...

Piotr Strzyzewski wrote:

...

On Mon, May 18, 2015 at 05:42:59PM +0200, Daniel Suchy wrote: [...] From my point of view, there was no formal proposal to discuss. Just an informal question by Eric. It is even unclear which objects are expected to be visible. IIRC there were a couple of quite informal discussions about this Q, some of them maybe even in the hallways :-) I have to admit that I deliberately not mentioned them, since they were (as you wrote) informal. I do believe, that interested parties should move those informal discussions to this mailing list. I distinctly remember that we were asked if we wanted to have the deleted objects viewable. In fact it was at RIPE 66 in Dublin. I'd like to think

Den 2015-05-19 14:44, Piotr Strzyzewski skrev: that this is formal enough to proceed to next level.

Slide 14. https://ripe66.ripe.net/presentations/186-DB_Update-66.pdf

13m30s in https://ripe66.ripe.net/archives/video/1174/

I asked about it after RIPE 67. https://www.ripe.net/ripe/mail/archives/db-wg/2013-November/004193.html And the answer was:

"There was no discussion or agreement on taking this any further at RIPE 67 or since on the mailing list. If the community would like us to provide this service, we can add this to our feature list for a future release of the RIPE Database software."

Since there was almost lack of discussion, my personal proposal is to first ask NCC legal department for legal comment about deleted objects being viewable. I would suggest limiting the list of possible objects to the same for which DB provides --show-version and --list-versions nowadays.

Lack of discussion is not a no. It's just lack of discussion. I said to Denis at RIPE 66 that I'd like to see this. I'm patient so I waited to next meeting. Nothing from RIPE NCC. So I asked on the mailing list and got the answer above. Which is not really an answer on the question that was put forward to us. But we can take it again. Here's the question from the presentation: "No deleted objects shown, should we show them?" My answer is "yes". I want this. Anyone else?

Is it ok with you and the rest of the WG?

Well. I'm not sure why we should bring in the NCC legal department. It's just a database with objects. /bengan -- Bengt Gördén Resilans AB

Den 2015-05-20 20:19, Piotr Strzyzewski skrev:

On Wed, May 20, 2015 at 08:04:51PM +0200, Bengt Gördén wrote:

...

...

Is it ok with you and the rest of the WG? Well. I'm not sure why we should bring in the NCC legal department. It's just a database with objects. Read the comment made by Wilfried Woeber about Data Protection.

Piotr

I did. I just happen to not agree with it. But lets not dwell on that. If you and Wilfried Woeber (or anyone else for that matter) think it's needed please do that. But rather sooner than later. When can we expect an answer from RIPE NCC legal department? /bengan

Hello, that's good point. But I have another - we still have major parts of RIPE database on FTP - ftp://ftp.ripe.net/ripe/dbase/ updated on regular basis. If someone is smart enough to mirror these (public) data periodically (and anyone can do that), he's also possible to analyse history (with all "deleted" data). Personal data aren't on FTP already - I think this is enough also for history publishing from RIPE DB history - as data protection is about *personal* data, in my oppinion. Historical data should ommit these pure personal details, but other (technical) history can be published. With regars, Daniel On 19.5.2015 12:45, Wilfried Woeber wrote:

The Data Protection environment states that an organisation is not supposed to collect data, or to keep storing it (and even less so, making it publicly accessible!) unless there is a need to do so for the declared reason, operation or goal.

As the obejcts have been deleted, it can be inferred that there is no longer a valid reason to store those items or to make them accessible.

Daniel, all let me be very clear about one thing: personally, I would also like to see the history of resource allocations. But, there's (on top of the PII Data Protection) the contractual stuff and the AUP. If there's no longer a Service Contract in place and an object is (voluntarily) deleted or the registration is removed by the RIR, then the question may be raised: ° is that information still to be kept and made publicly available? That's the reason why I think we should seek both: - guidance or request by the community, and if requested (or maybe in parallel with the discussion) - a review by the NCC's Legal Department. Wilfried On 2015-05-20 22:32, Daniel Suchy wrote:

Hello, that's good point. But I have another - we still have major parts of RIPE database on FTP - ftp://ftp.ripe.net/ripe/dbase/ updated on regular basis.

If someone is smart enough to mirror these (public) data periodically (and anyone can do that), he's also possible to analyse history (with all "deleted" data).

Personal data aren't on FTP already - I think this is enough also for history publishing from RIPE DB history - as data protection is about *personal* data, in my oppinion. Historical data should ommit these pure personal details, but other (technical) history can be published.

With regars, Daniel

On 19.5.2015 12:45, Wilfried Woeber wrote:

...

The Data Protection environment states that an organisation is not supposed to collect data, or to keep storing it (and even less so, making it publicly accessible!) unless there is a need to do so for the declared reason, operation or goal.

As the obejcts have been deleted, it can be inferred that there is no longer a valid reason to store those items or to make them accessible.

On Thu, May 21, 2015 at 11:37:35AM +0200, Wilfried Woeber wrote: Dear DB WG Members

let me be very clear about one thing: personally, I would also like to see the history of resource allocations.

Just to be clear - me too. Piotr -- gucio -> Piotr Strzyżewski E-mail: Piotr.Strzyzewski@polsl.pl