Re: fuzzy subnet aggregation

[Joe Maimon]

So I went back to the drawing board, and I think I have something that 
seems to work much better.


- convert input prefixes to single ip expressed as integer
- sort -n | uniq
- into a temporary list file

begin

read sequentially until maxhosts (or minhosts) or next subnet

If matched enough single addresses, output subnet (and missing hosts 
without early loop termination)


delete all subnet addresses read

loop

Total process time on a vm on old hardware, less than 2m for a 5500 line 
input. Now to verify results, positive and negative


Results are still raw, but anyone who wishes is welcome to it.

Joe

Joe Maimon wrote:

Does anyone have or seen any such tool? I have a script that seems to
work, but its terribly slow.

Currently I can produce aggregated subnets that can be mising up to a
specified number of individual addresses. Which can be fed back in for
multiple passes.

Doing RTBH on individual /32 does not scale well, if you are eyeing
collaboration with external lists. I have found likely sources that could
produce another 100k prefixes easily.

Joe

Re: California network infrastructure report (FCC)

[Tim Požár]

That lines up with my experience living in Mill Valley and trying to
tether here.  Either cell service is down, or it is up but not data.
Finally gave up and drove into SF to steal some electrons and wifi at a
dive bar on Geary.

Tim

On 10/27/19 5:21 PM, Sean Donelan wrote:
> 
> According to reporting to the FCC:
> https://www.fcc.gov/document/ca-power-shutoff-communications-status-report-oct-27-2019
> 
> 
> Cell sites out of service: overall 2.4% (630 out of 25,893)
> 
> Marin County: 49.6% out of service (105 out of 270)
> Lake County: 19.3% out of service (11 out of 57)
> Calaveras County: 18.6% out of service (8 out of 43)
> Sonoma County: 17.1% out of service (73 out of 427)
> Santa Cruz County: 16.4% out of service (35 out of 213)
> 
> Cable and Wireline systems (combined)
> 
> 393,735 subscribers out of service. Only Outside Plant and Hub
> locations. Likely does not include loss of Customer Premise Equipment
> (CPE) power.
> 
> 
> Broadcast
> 
> 4 FM radio stations out of service, 2 FM stations with programming sent
> to another station.
> 

Re: fuzzy subnet aggregation

[Antonio Querubin]

Are you trying to reduce the number of ACL rules that include a known set of 
addresses but also minimize covered addresses that are not part of the 
mandatory set?

Tony

> On Oct 27, 2019, at 12:29, Joe Maimon  wrote:
> 
> 
>> 
>> On Sun, Oct 27, 2019 at 3:09 PM Joe Maimon  wrote:
> 
>>> 
>> 
>> your aim is to get to maximum aggregation .. with some overage, like
>> 90% of a /24  ?
>> so missing like 25 addresses in a whole /24.. (for instance)
> 
> I would be happy to get /29's missing 3 /28's missing 5, etc...
> 
> This is not punitive, its about scale.
> 
> Joe
> 

RE: IPv4 and Auctions

[Michel Py]

> John Curran wrote :
>  So, if by “the right to use them”, one is referring to being the one listed 
> in the ARIN database for the address space and/or use ARIN services applicable
> to those address blocks, then that is indeed a contractual right, but it 
> doesn’t get transferred or assigned except as the community policy states.  
> For
>  example, redelegation by ISPs is clearly covered by ARIN policy, so we 
> recognize such and even provide services specifically to support same.

I wish to retract what I wrote earlier : I totally acknowledge and support 
ARIN's phrasing, and it was absolutely not meant as a challenge.
ARIN does not lease me the IP block I thought I "bought" on the transfer market.

What I "bought" is, by the result of a complex by-product iterations of ARIN 
acknowledgement that although I neither own or rent the apartment, my name is 
on it, I somehow expect that the community will somehow consider that my org is 
the one who should announce the prefix allocated / assigned to said org, and 
not someone else.

Is this appropriate langage ?

I am going to move the thread of transparency to arin-ppml, where it belongs.

Michel.

TSI Disclaimer:  This message and any files or text attached to it are intended 
only for the recipients named above and contain information that may be 
confidential or privileged. If you are not the intended recipient, you must not 
forward, copy, use or otherwise disclose this communication or the information 
contained herein. In the event you have received this message in error, please 
notify the sender immediately by replying to this message, and then delete all 
copies of it from your system. Thank you!...

Re: U.S. Senate: READI Act 2019 re-introducted

[Masataka Ohta]

Sean Donelan wrote:

Somehow, I doubt if their recent experience with wildfires in the Bay 
Area will change any tech CEOs' opinions or amazon, google, apple smart 
product managers' project plans for supporting emergency alerts.


There already exists earthquake alert system in CA.

Masataka Ohta

California network infrastructure report (FCC)

[Sean Donelan]

According to reporting to the FCC:
https://www.fcc.gov/document/ca-power-shutoff-communications-status-report-oct-27-2019

Cell sites out of service: overall 2.4% (630 out of 25,893)

Marin County: 49.6% out of service (105 out of 270)
Lake County: 19.3% out of service (11 out of 57)
Calaveras County: 18.6% out of service (8 out of 43)
Sonoma County: 17.1% out of service (73 out of 427)
Santa Cruz County: 16.4% out of service (35 out of 213)

Cable and Wireline systems (combined)

393,735 subscribers out of service. Only Outside Plant and Hub locations. 
Likely does not include loss of Customer Premise Equipment (CPE) power.



Broadcast

4 FM radio stations out of service, 2 FM stations with programming sent to 
another station.

Re: U.S. Senate: READI Act 2019 re-introducted

[Michael Thomas]

On 10/27/19 4:46 PM, Sean Donelan wrote:

On Sun, 27 Oct 2019, Sean Donelan wrote:
I do not expect Apple, Amazon or Google to do something until forced 
too.




The semi-joke amoung the emergency management community, if tech firm 
CEOs lived in the mid-west (tornado alley) or south-east (hurricane 
coasts) instead of west-coast (silicon valley & seattle), all tech 
products would already support emergency alerts.


Somehow, I doubt if their recent experience with wildfires in the Bay 
Area will change any tech CEOs' opinions or amazon, google, apple 
smart product managers' project plans for supporting emergency alerts.


Samsung is likely the most advanced in this area, because the South 
Korean government has been 'encouraging' korean firms for several 
years to build emergency alert technology into their products.



In other news, Generac has been doing land office business lately.

https://www.cnbc.com/2019/10/25/generac-soars-to-a-record-as-pge-power-cuts-cause-business-to-boom.html

Mike

Re: U.S. Senate: READI Act 2019 re-introducted

[Michael Thomas]

On 10/27/19 4:46 PM, Sean Donelan wrote:

On Sun, 27 Oct 2019, Sean Donelan wrote:
I do not expect Apple, Amazon or Google to do something until forced 
too.




The semi-joke amoung the emergency management community, if tech firm 
CEOs lived in the mid-west (tornado alley) or south-east (hurricane 
coasts) instead of west-coast (silicon valley & seattle), all tech 
products would already support emergency alerts.


Somehow, I doubt if their recent experience with wildfires in the Bay 
Area will change any tech CEOs' opinions or amazon, google, apple 
smart product managers' project plans for supporting emergency alerts.


Samsung is likely the most advanced in this area, because the South 
Korean government has been 'encouraging' korean firms for several 
years to build emergency alert technology into their products.



Considering that most of the ritzy areas in the silly valley have no 
power right now, this may take on a new urgency. Of course none of this 
really matters if the network infrastructure isn't backed up. For all of 
the money sloshing around the valley, we have really shitty network 
infrastructure.


Mike

Re: U.S. Senate: READI Act 2019 re-introducted

[Sean Donelan]

On Sun, 27 Oct 2019, Sean Donelan wrote:

I do not expect Apple, Amazon or Google to do something until forced too.



The semi-joke amoung the emergency management community, if tech firm CEOs 
lived in the mid-west (tornado alley) or south-east (hurricane coasts) 
instead of west-coast (silicon valley & seattle), all tech products would 
already support emergency alerts.


Somehow, I doubt if their recent experience with wildfires in the Bay Area 
will change any tech CEOs' opinions or amazon, google, apple smart product 
managers' project plans for supporting emergency alerts.


Samsung is likely the most advanced in this area, because the South Korean 
government has been 'encouraging' korean firms for several years to build 
emergency alert technology into their products.

Re: fuzzy subnet aggregation

[Masataka Ohta]

Joe Maimon wrote:


Does anyone have or seen any such tool? I have a script that seems to
work, but its terribly slow.


It's a logic synthesis problem and should be NP hard.

Masataka  Ohta

Re: fuzzy subnet aggregation

[Grant Taylor via NANOG]

On 10/27/19 4:27 PM, Joe Maimon wrote:

I would be happy to get /29's missing 3 /28's missing 5, etc...


Are you good with rounding up to the next larger network if you have 
~62% of the members?



This is not punitive, its about scale.


ACK



--
Grant. . . .
unix || die



smime.p7s
Description: S/MIME Cryptographic Signature

Re: IPv4 and Auctions

[John Curran]

On 26 Oct 2019, at 8:28 AM, Owen DeLong 
mailto:o...@delong.com>> wrote:
…
The difference is that ARIN charges almost nothing for the rent, so what you 
basically are auctioning is the right too use a free appartment, which is worth 
money.
Even if you don't own the IP addresses, the right to use them is a tangible 
asset.

I’m sure someone from ARIN staff will correct me if I’m wrong about this, but I 
don’t believe that is the case.

ARIN registration is NOT (to the best of my knowledge):
+ A right to use
+ A property right in a set of integers
+ Any sort of right to have your numbers routed on the internet (or anywhere)
+ Any sort of exclusive right to control of a set of integers for a particular 
purpose

ARIN registration is (to the best of my knowledge):
+ A guarantee of unique association of a set of integers to your organization 
within a
cooperating set of databases collectively known as the RIR System.
+ A guarantee of certain property and control rights over said registration 
within that
system. (note that’s the registration, not the registered integers)
+ Unless you are a non-RSA legacy registrant, it is a contractual relationship 
between
you and ARIN (and by extension said RIR system) which provides both rights and
obligations on your conduct with respect to said registration.
+ The right and ability to update certain attributes in the records of your 
registration(s).

So ARIN doesn’t actually rent the right to use an apartment so much as a 
recording of the fact
that certain entities agree that your name goes on the door of said apartment.

Owen -

Actually, that’s quite close.  To be clear on this, I’ll point out that ARIN 
recognizes that those issued IP address blocks have several specific rights –

• The exclusive right to be the registrant of the resources within the ARIN 
database
• The right to use the number resources _within the ARIN database_
• The right to transfer the number resources pursuant to the community’s 
policies.

These rights are provided contractually to all parties with ARIN-issued 
resources, and ARIN will recognize and formalize the rights of those issued 
resources by ARIN’s predecessors (legacy resource holders) by entering into a 
registration services agreement with them if wish clear contractual rights over 
their resources.

Note that these rights cannot be assigned or transferred without ARIN’s consent 
and such consent may not be unreasonably withheld if consistent with the 
policies.

So, if by “the right to use them”, one is referring to being the one listed in 
the ARIN database for the address space and/or use ARIN services applicable to 
those address blocks, then that is indeed a contractual right, but it doesn’t 
get transferred or assigned except as the community policy states.  For 
example, redelegation by ISPs is clearly covered by ARIN policy, so we 
recognize such and even provide services specifically to support same.

If “the right to use them” is rather a reference to ability to route address 
blocks with your various ISP partners, then that’s really a question about the 
business practices of those accepting the routes…

Now, coincidentally and fortunately, the vast majority of ISPs choose to regard 
the data in the
RIR system as an important record of who they will accept prefix advertisements 
from, which
makes it much harder to use numbers that are not associated with your 
organization in the RIR system
for routing on the internet, but that’s actually a coincidental behavior of the 
ISPs and not actually
any sort of right, privilege, or ability issued or managed by ARIN.

Correct.   ARIN’s policies govern the administration of the number resources in 
the registry, and there is no requirement for resource holders to route their 
networks in any particular manner.   During the Anti-hijack policy discussion 
on arin-p...@arin.net,  I noted that if the 
community really wanted ARIN to require certain routing hygiene, that would 
require changing the RSA, and any changes to ARIN’s RSA going forward (outside 
of conformance to changing law) actually requires a member ratification vote…  
(a particularly high hurdle, but potentially achievable if the community really 
feels that they want additional obligations in this regard.)

Thanks,
/John

John Curran
President and CEO
American Registry for Internet Numbers

Re: fuzzy subnet aggregation

[Joe Maimon]

> On Sun, Oct 27, 2019 at 3:09 PM Joe Maimon  wrote:

>>
>
> your aim is to get to maximum aggregation .. with some overage, like
> 90% of a /24  ?
> so missing like 25 addresses in a whole /24.. (for instance)

I would be happy to get /29's missing 3 /28's missing 5, etc...

This is not punitive, its about scale.

Joe

Re: fuzzy subnet aggregation

[Mark Leonard]

Is this what you are trying to accomplish?

$ python
Python 2.7.15rc1 (default, Nov 12 2018, 14:31:15)
[GCC 7.3.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import netaddr
>>> SomeList=netaddr.IPSet()
>>> SomeList.add('203.0.113.0/25')
>>> SomeList.add('203.0.113.128/25')
>>> for x in list(SomeList.iter_cidrs()):
...   print x
...
203.0.113.0/24
>>>
>>> DifferentList=netaddr.IPSet()
>>> DifferentList.add('0.0.0.0/0')
>>> DifferentList.remove('203.0.113.1')
>>> for x in list(DifferentList.iter_cidrs()):
...   print x
...
0.0.0.0/1
128.0.0.0/2
192.0.0.0/5
200.0.0.0/7
202.0.0.0/8
203.0.0.0/18
203.0.64.0/19
203.0.96.0/20
203.0.112.0/24
203.0.113.0/32
203.0.113.2/31
203.0.113.4/30
203.0.113.8/29
203.0.113.16/28
203.0.113.32/27
203.0.113.64/26
203.0.113.128/25
203.0.114.0/23
203.0.116.0/22
203.0.120.0/21
203.0.128.0/17
203.1.0.0/16
203.2.0.0/15
203.4.0.0/14
203.8.0.0/13
203.16.0.0/12
203.32.0.0/11
203.64.0.0/10
203.128.0.0/9
204.0.0.0/6
208.0.0.0/4
224.0.0.0/3
>>>

On Sun, Oct 27, 2019 at 1:10 PM Joe Maimon  wrote:

> Does anyone have or seen any such tool? I have a script that seems to
> work, but its terribly slow.
>
> Currently I can produce aggregated subnets that can be mising up to a
> specified number of individual addresses. Which can be fed back in for
> multiple passes.
>
> Doing RTBH on individual /32 does not scale well, if you are eyeing
> collaboration with external lists. I have found likely sources that could
> produce another 100k prefixes easily.
>
> Joe
>

Re: U.S. Senate: READI Act 2019 re-introducted

[Sean Donelan]

Number of apps available in leading app stores 2019

Google Play: 2,470,000
Apple App Store: 1,800,000
Windows Store: 669,000
Amazon Appstore: 487,000

Likely hood all, a majority, a minority or even a tiny percentage of App 
developers will do the right thing?  Close to zero.  How many Apps are 
written in other countries, and don't follow requirements across borders.


Its not even a hypothetical, we know this from experience with cellular 
telephones in the early 2000s.  The cellular industry and mobile device 
OEMs claimed for 10 years that they should NOT supply emergency alerts, 
because all the Apps would do that.  The cell phone was "just the platform"


Didn't happen. There were a few, very few apps, which implemented alerts; 
but default settings is a powerful thing.  Almost none of the public 
installed or used them.  Even when cell telephone companies 
isntalled alert apps by default, they failed to maintain them and 
often didn't work when needed.


Eventually, Congress passed the AWARN legislation requiring cell phone 
operators and manufacturers to implement emergency alerts.  A subscriber 
can opt-out, but by default all cellular telephone OEMs and OSs must 
implement emergency alerts.  A decade later, Netflix, Hulu, Spotify or 
whatever App you are using on your phone still rarely implement alerts. 
Wireless Emergency Alerts are nearly always triggered by the base cell 
phone operating system.


Amazon Alexa (echo operating system), Google Assistant (Google home/nest 
operating system), etc. are avoiding it much like the old cell phone OEMs 
in the mid-2000s.


But eventually I expect there will be a disaster, and lots of people 
won't get warnings, and will die.  Cable TV operators fought implementing 
EBS/EAS through the 1980s. Cable TV didn't have EBS/EAS, and several 
hundred people died watching premium cable in the midwest and didn't get 
the tornado warnings being broadcast by the local TV stations.  A few 
years later, Congress passed the law requiring Cable TV operators to 
implement EAS/EBS.


Much like seatbelts and car manufacturers, I expect tech firms to dodge as 
long as possible.


I appreciate your belief that somehow industry will do the right thing on 
its own.  History isn't on your side.


I do not expect Apple, Amazon or Google to do something until forced too.

Re: fuzzy subnet aggregation

[Christopher Morrow]

On Sun, Oct 27, 2019 at 3:09 PM Joe Maimon  wrote:
>
> Does anyone have or seen any such tool? I have a script that seems to
> work, but its terribly slow.
>
> Currently I can produce aggregated subnets that can be mising up to a
> specified number of individual addresses. Which can be fed back in for
> multiple passes.
>

your aim is to get to maximum aggregation .. with some overage, like
90% of a /24  ?
so missing like 25 addresses in a whole /24.. (for instance)

> Doing RTBH on individual /32 does not scale well, if you are eyeing
> collaboration with external lists. I have found likely sources that could
> produce another 100k prefixes easily.
>
> Joe

Re: IPv4 and Auctions

[Owen DeLong]

> On Oct 27, 2019, at 09:47 , Michel Py  wrote:
> 
> Hi Owen,
> 
>> Owen DeLong wrote :
>> So ARIN doesn’t actually rent the right to use an apartment so much as a 
>> recording of the fact
>> that certain entities agree that your name goes on the door of said 
>> apartment.
> 
> Correct, but in the end I still have the apartment for very cheap, good 
> enough for me.
> If a squatter tries to use it, good chances are that community efforts, not 
> ARIN because ARIN is not the police, will lead to me retaining the use of it.
> 
> 
>>> Michel Py wrote :
>>> What I like with Hilco is that it brings transparency to the market. I 
>>> think that each transfer should list the amount of the
>>> transaction between parties. For example, I would like to know for how much 
>>> 44.192/10 went.
> 
>> Owen DeLong wrote :
>> If you really feel that this should be data the RIRs collect during 
>> transfers and that it should be published, I suggest you submit a proposal
>> for this into the ARIN policy development process. If you need help doing 
>> so, feel free to ask me or any other member of the AC.
> 
> I think the result should be simple, a .csv file containing an entry for each 
> prefix transferred :
> Date, size, price, origin RIR, resulting RIR.
> Something like https://auctions.ipv4.global/prior-sales but covering all 
> transactions, not only ipv4.global ones.
> Transparency on transfer prices.
> 
> What do you think about it ? a two-prong question :

I’m not super enthusiastic about it one way or the other, to be honest. I 
wouldn’t oppose a policy favoring market transparency,
but I doubt it will achieve significant support from the rest of the community 
and I’m sure there will be vocal opposition.

> - As yourself ? is it desirable for the community  in your opinion ?

I think it could offer some benefit to the community. OTOH, I can also 
understand why some entities prefer to keep their financial
data private. I’m honestly somewhat ambivalent about where the balance should 
be struck here between privacy and transparency.

> - As AC member ? Does it have any chance to be approved by the AC ?

That’s not really a valid question, IMHO. I think anything which garners 
significant support in the community has a chance to be
approved by the AC. We are representatives of the community. I can’t speak for 
other AC members, but I will say that I do sometimes
find myself voting on the AC in a manner that does not reflect my personal 
views because I have seen clear indication from the
community that the will of the community differs from my personal opinion.

> I would submit a proposal if it has some chances to pass; I don't want to 
> lose the time of the AC if it's going to be deep-sixed right away.

The only way a proposal gets deep-sixed by the AC right away is if it is out of 
scope or has an unclear problem statement.

After that, the AC only abandons proposals that have received significant 
opposition from the community and which the AC does not believe
have will ever reach consensus within the community.

> Side question more towards John or the ARIN staff, how much work is that 
> going to be to implement ?

Some questions likely necessary for John or staff to answer:

Which classes of transfers would you apply this to?
8.2 (M)?
8.3 (Specified Transfers within the ARIN region)
8.4 (Inter-RIR transfers to/from ARIN)
Inbound only?
Outbound only?
Both?

What about 8.4 transfers where the party not in the ARIN region insists on a 
confidentiality agreement about pricing? How far
do you want to extend the ARIN policy into affecting the behavior of suppliers 
and/or recipients in other regions?

Owen

> 
> Michel.
> 
> TSI Disclaimer:  This message and any files or text attached to it are 
> intended only for the recipients named above and contain information that may 
> be confidential or privileged. If you are not the intended recipient, you 
> must not forward, copy, use or otherwise disclose this communication or the 
> information contained herein. In the event you have received this message in 
> error, please notify the sender immediately by replying to this message, and 
> then delete all copies of it from your system. Thank you!...

Re: U.S. Senate: READI Act 2019 re-introducted

[Michael Thomas]

On 10/27/19 11:57 AM, Sean Donelan wrote:

On Fri, 25 Oct 2019, Michael Thomas wrote:
Ok, you had me completely puzzled by digital assistant layer. I'm not 
sure apps might not be interested in competing for users: "This 7.0 
earthquake is brought to you by Allstate!"


I'll assume you intended a smiley emoticon.



Yes. My real issue is that I'm not sure I want the USG in the business 
of UI requirements, and I'm not sure that I want Apple and Google to be 
the sole arbiters either. Take for example Amber alerts: their ham 
fisted blasts that don't even take into account whether I'm in any 
position to help... like I'm sitting at home and not on the road. And 
then there's the issue of 8 dozen devices in the house faithfully 
blasting the same message. And then of course, there is the 
disaggregation problem of many different sources of altering using 
different alert distribution mechanisms. It would be nice to have some 
leeway such that somebody who might care about those problems has a 
chance at to be a player too. Yes, maybe that is an Allstate app that I 
can opt into. Maybe it's an open source project. Maybe it needs 
certification. There are a lot of possibilities here and although Apple 
and Google are at approximately the right layer in the stack, I don't 
think they're at the right layer of motivation: Allstate would 
definitely have more motivation, IMO.


Mike

Re: fuzzy subnet aggregation

[Nick Morrison via NANOG]

> On 27. Oct 2019, at 20:36, Joe Maimon  wrote:
> 
> Not quite.
> 
> 203.0.113.1
> 203.0.113.3
> 203.0.113.5
> 203.0.112.6
> 203.0.112.7
> 
> Will aggregate to 203.0.113.0/29 if you dont mind the missing 3 addresses
> in the unaggregated list.
> 
> Hence, fuzzy aggregation.

Could you describe the problem again? I’m interested, but I’m not sure that I 
quite understand what you want to do :-) were the last two addresses supposed 
to have 112 in the third octet?

Nick

Re: fuzzy subnet aggregation

[Joe Maimon]

Not quite.

203.0.113.1
203.0.113.3
203.0.113.5
203.0.112.6
203.0.112.7

Will aggregate to 203.0.113.0/29 if you dont mind the missing 3 addresses
in the unaggregated list.

Hence, fuzzy aggregation.

Joe

> Is this what you are trying to accomplish?
>
> $ python
> Python 2.7.15rc1 (default, Nov 12 2018, 14:31:15)
> [GCC 7.3.0] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
 import netaddr
 SomeList=netaddr.IPSet()
 SomeList.add('203.0.113.0/25')
 SomeList.add('203.0.113.128/25')
 for x in list(SomeList.iter_cidrs()):
> ...   print x
> ...
> 203.0.113.0/24

 DifferentList=netaddr.IPSet()
 DifferentList.add('0.0.0.0/0')
 DifferentList.remove('203.0.113.1')
 for x in list(DifferentList.iter_cidrs()):
> ...   print x
> ...
> 0.0.0.0/1
> 128.0.0.0/2
> 192.0.0.0/5
> 200.0.0.0/7
> 202.0.0.0/8
> 203.0.0.0/18
> 203.0.64.0/19
> 203.0.96.0/20
> 203.0.112.0/24
> 203.0.113.0/32
> 203.0.113.2/31
> 203.0.113.4/30
> 203.0.113.8/29
> 203.0.113.16/28
> 203.0.113.32/27
> 203.0.113.64/26
> 203.0.113.128/25
> 203.0.114.0/23
> 203.0.116.0/22
> 203.0.120.0/21
> 203.0.128.0/17
> 203.1.0.0/16
> 203.2.0.0/15
> 203.4.0.0/14
> 203.8.0.0/13
> 203.16.0.0/12
> 203.32.0.0/11
> 203.64.0.0/10
> 203.128.0.0/9
> 204.0.0.0/6
> 208.0.0.0/4
> 224.0.0.0/3

>
> On Sun, Oct 27, 2019 at 1:10 PM Joe Maimon  wrote:
>
>> Does anyone have or seen any such tool? I have a script that seems to
>> work, but its terribly slow.
>>
>> Currently I can produce aggregated subnets that can be mising up to a
>> specified number of individual addresses. Which can be fed back in for
>> multiple passes.
>>
>> Doing RTBH on individual /32 does not scale well, if you are eyeing
>> collaboration with external lists. I have found likely sources that
>> could
>> produce another 100k prefixes easily.
>>
>> Joe
>>
>

Re: Couple of questions about "baremetal/ONIE" networking equipment sellers

[Tore Anderson]

* Nick ten Cate

> We also have lots of experience with FS.com switches; however.. One thing we 
> noticed really quick is that its better to order 1 and to find the actual 
> supplier and order with them directly. FS.com is a reseller; and they will 
> switch (no pun intended) supplier almost yearly. Real technical support is 
> nonexistent (even though they claim it is great) and I have yet to have a 
> single bug fixed; packet dumps and steps to reproduce included. I have 
> removed all of our *N*5850-48S6Q due to bugs in software lockups.

Hi Nick,

FS.com did indeed replace their N5850-58S6Q supplier a while back. It is rather 
idiotic of them to not change their SKU when they do so.

Anyway, before it was manufactured by Celestica I think, now it is the 
Edge-Core AS5812-54X. The latter is very well supported by Cumulus, the former 
is not.

You can see it is the the Edge-Core by comparing the pictures:

https://www.fs.com/de-en/products/69226.html
https://www.edge-core.com/productsInfo.php?cls=1=8=59=119

We bought a few of them. I did mail our AM before placing the order to 
ascertain that they would indeed deliver the AS5812-54X and to make it crystal 
clear that no other model would be accepted. No problem.

They will also sell other Edge-Core models that's not (yet) on their website 
catalogue if you ask (we ordered a few AS7326-56Xes).

I do not believe Edge-Core will sell direct to end-users, so resellers like FS, 
Cumulus Networks or HPE is your best bet if you want those.

Tore

fuzzy subnet aggregation

[Joe Maimon]

Does anyone have or seen any such tool? I have a script that seems to
work, but its terribly slow.

Currently I can produce aggregated subnets that can be mising up to a
specified number of individual addresses. Which can be fed back in for
multiple passes.

Doing RTBH on individual /32 does not scale well, if you are eyeing
collaboration with external lists. I have found likely sources that could
produce another 100k prefixes easily.

Joe

Re: U.S. Senate: READI Act 2019 re-introducted

[Sean Donelan]

On Fri, 25 Oct 2019, Michael Thomas wrote:
Ok, you had me completely puzzled by digital assistant layer. I'm not sure 
apps might not be interested in competing for users: "This 7.0 earthquake is 
brought to you by Allstate!"


I'll assume you intended a smiley emoticon.

Do not use interstitials, ad pre-rolls, captchas, etc during actual 
emergency alert information.


Since new people seem to propose it periodically, it turns out advertisers 
(and consumers) do not like their brands being associated with mass 
casualty events, child abductions and terrorism incidents.  High-quality 
(i.e. high-revenue) marketeers demand buffers between their ads and 
sensitive topics to avoid being branded explotive. That's why you don't 
see airline advertising for days or sometimes weeks after a major 
airplane crash.


Radio and television have learned this lesson over decades. The Weather 
Channel is very good at keeping ads separate from actual alerts. Even 
algorithmic and auction-based on-line advertising and social media 
networks are mostly learning this lesson, usually the hard way.


After the immediate disaster, marketeers do use geo-targeting. But even 
then, the better advertising agencies change their messaging in disaster 
areas.


https://adage.com/article/digitalnext/advertising-disaster-regions/310389
https://www.nytimes.com/2012/11/01/business/media/marketers-ride-the-coattails-of-a-storm-not-all-successfully.html


Finally, the FCC has been fining advertisers over $1 million for using 
official emergency alert tones and signals in ads to get people's 
attention.


The techies in silicon valley should learn from their marketeering
counter-parts on madison avenue -- keep your emergency alerts separate 
from your advertising.

RE: IPv4 and Auctions

[Michel Py]

Hi Owen,

> Owen DeLong wrote :
> So ARIN doesn’t actually rent the right to use an apartment so much as a 
> recording of the fact
> that certain entities agree that your name goes on the door of said apartment.

Correct, but in the end I still have the apartment for very cheap, good enough 
for me.
If a squatter tries to use it, good chances are that community efforts, not 
ARIN because ARIN is not the police, will lead to me retaining the use of it.


>> Michel Py wrote :
>> What I like with Hilco is that it brings transparency to the market. I think 
>> that each transfer should list the amount of the
>> transaction between parties. For example, I would like to know for how much 
>> 44.192/10 went.

> Owen DeLong wrote :
> If you really feel that this should be data the RIRs collect during transfers 
> and that it should be published, I suggest you submit a proposal
> for this into the ARIN policy development process. If you need help doing so, 
> feel free to ask me or any other member of the AC.

I think the result should be simple, a .csv file containing an entry for each 
prefix transferred :
Date, size, price, origin RIR, resulting RIR.
Something like https://auctions.ipv4.global/prior-sales but covering all 
transactions, not only ipv4.global ones.
Transparency on transfer prices.

What do you think about it ? a two-prong question :

- As yourself ? is it desirable for the community  in your opinion ?

- As AC member ? Does it have any chance to be approved by the AC ?

I would submit a proposal if it has some chances to pass; I don't want to lose 
the time of the AC if it's going to be deep-sixed right away.

Side question more towards John or the ARIN staff, how much work is that going 
to be to implement ?

Michel.

TSI Disclaimer:  This message and any files or text attached to it are intended 
only for the recipients named above and contain information that may be 
confidential or privileged. If you are not the intended recipient, you must not 
forward, copy, use or otherwise disclose this communication or the information 
contained herein. In the event you have received this message in error, please 
notify the sender immediately by replying to this message, and then delete all 
copies of it from your system. Thank you!...