Dear Cedrick,

It is ok with me if you use one of the 2 VMs in Dar es Salaam for auth
DNS to also do the digging for the lameness, ie as a second node.

I trust these are sufficiently diverse from your 1st node, they are in
AS327844, with upstreams AS37084 and AS30844.

Hope that can help.

Regards,
Frank


On 10/09/2020 21:20, Cedrick Adrien Mbeyet wrote:
> Dear dbwg,
> 
> 
> Referring to the previous email from my colleague Simon.
> 
> We indeed delayed the deletion of the lame records for multiple reason
> among them the absence of a second nodes. We had some security
> challenges that needed to be addressed before being able to plan the
> second nodes.
> 
> We do apologize for the delay on the second node implementation and rest
> tune as deployment will be scheduled very soon. And usual, announced on
> the usual channels.
> 
> Thanks and regards,
> 
> 
> -- 
> _______________________________________________________________
> Cedrick Adrien Mbeyet                                           
> IT Infrastructure Unit Manager, AFRINIC Ltd.
> t:  +230 403 5100 / 403 5115 | f: +230 466 6758 | tt: @afrinic | w: 
> www.afrinic.net
> facebook.com/afrinic | flickr.com/afrinic | youtube.com/afrinicmedia
> ______________________________________________________
> 
> On 07/09/2020 16:59, Simon Seruyinda wrote:
>> Hi Frank,
>>
>> The e-mail attribute was made mandatory in July 2012.
>> I have done a quick check in the database and we have 1048 person objects 
>> without the email attribute.
>> Most of these objects belong to legacy resource holders and were imported 
>> into the database during the initial setup.
>> Many are referenced in different objects. Below are some stats regarding 
>> number of objects that are referencing these person objects as 
>> admin-c,tech-c or zone-c.
>>
>> zone-c:
>> ===============
>> 51 domain objects
>>
>> tech-c
>> ===============
>> as-block     248
>> as-set       11
>> domain       35
>> inetnum      574
>> mntner       60
>> org  162
>> role 7
>> route-set    6
>>
>> admin-c:
>> ==============
>> as-block     248
>> as-set       11
>> aut-num      2
>> domain       49
>> inet6num     1
>> inetnum      731
>> mntner       71
>> org  137
>> role 4
>> route-set    6
>>
>> There is an ongoing project internally focused on contacting these legacy 
>> holders in order to update their contact details in the database. Another 
>> activity, under the scope of the whois business rules inconsistencies is 
>> also planned to get the emails updated for any resource members who may be 
>> having no emails in the any of their person objects. Incases where efforts 
>> to get in touch with the resource holder proves futile, a temporary measure 
>> using AFRINIC’s placeholder email accounts is undertaken. These activities 
>> are expected to decrease the number significantly.
>>
>> With regards to the lame delegation handling, we are not doing deletion yet 
>> since we are running only one node to do the lame delegation checks. Once 
>> the second node is setup, we shall begin the deletion otherwise for now we 
>> run the risk of a few false positives.
>>
>> Regarding the rdns objects size, thanks for bringing this up for discussion. 
>> Currently we have a limit for IPv4 set to minimum of /24, but there is no 
>> limit implemented for IPv6, so it will go up to 128.
>> I agree this could lead to unnecessary db growth and i think a limit should 
>> be set. Input from the DBWG members on what would be the appropriate minimum 
>> would highly be appreciated.
>>
>> Regards;
>> Simon
>>
>>> On 6 Sep 2020, at 22:22, Frank Habicht <[email protected]> wrote:
>>>
>>> Hi AfriNIC staff,
>>>
>>> since when is the 'e-mail:' attribute for 'person' objects mandatory?
>>>
>>> I just found
>>> nic-hdl:        SE1-AFRINIC
>>> that does not have an email.
>>>
>>> It's got a GENERATED maintainer, and I'm also wondering how these new
>>> maintainer credentials were communicated to the "person".
>>>
>>> Yes, I don't want to rely on 'changed:' attributes.
>>>
>>> Staff:
>>> How many 'person' objects don't have an 'e-mail:' attribute ?
>>>
>>>
>>> [slowly getting to another issue....]
>>>
>>> Why did I get to check this person object at all....?
>>>
>>> Because in a domain object it is
>>> tech-c:         SE1-AFRINIC
>>> zone-c:         SE1-AFRINIC
>>>
>>>
>>> Also, the domain object is since "2020-02-02 02:02"
>>> ( nice time stamp!! ;-) ) marked as all 'nserver' being *lame*.
>>> So when is it meant to get deleted?
>>> I hope we're not waiting for the tech-c or zone-c to respond to the
>>> email, which we could not send, because the 'person' doesn't have an
>>> email address?
>>>
>>> But what really got me to check the domain object:
>>>
>>> domain:
>>> 0.1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.1.0.0.8.f.3.4.1.0.0.2.ip6.arpa
>>>
>>> yes, it's a bit long. a reverse DNS delegation for a /128
>>>
>>> This is probably "legal".
>>> But:
>>> a) if disputable 'usefulness', and
>>> b) I see "tremendous' potential for growth in the DB - in a bad way
>>>
>>>
>>> All, Staff and WG:
>>>
>>> should creation of domain objects be limited to certain prefix sizes?
>>>
>>>
>>> Thanks,
>>> Frank
>>>
>>> _______________________________________________
>>> DBWG mailing list
>>> [email protected]
>>> https://lists.afrinic.net/mailman/listinfo/dbwg
>> _______________________________________________
>> DBWG mailing list
>> [email protected]
>> https://lists.afrinic.net/mailman/listinfo/dbwg

_______________________________________________
DBWG mailing list
[email protected]
https://lists.afrinic.net/mailman/listinfo/dbwg

Reply via email to