October 2001 - lir-wg - mailman.ripe.net

Re: Multihoming - Resilience or Independence
by SchmitzJo＠aol.com 11 Oct '01

11 Oct '01

Dear Nurani, you wrote: > Rather than the RIRs giving advise on these operational matters, we think > it would be appropriate and helpful if these recommendations would come > from the ISP community. I fully support this statement. > A few BCPs or guideline documents would be useful to point these > requesters to. Could this be an initiative for the routing-wg? I had some general input on this in the past months. I am also aware that there are some very early drafts which had not been circulated widely. interesting in various ways ;-) I have no objections against discussing this in the Routing WG, even with the initial modest goal of collecting the individual points of view (which would form a suitable basis to further develop the ideas presented). Nonetheless, it is obvious that within the Routing WG the focus is on technical issues only. Anybody who is interested in summarizing and presenting his/hers ideas is welcome to do so. I can easily reserve some time for that in the next Routing WG session in January to give this a focus. Just step up... :-) Thanks Joachim --- JS395-RIPE -- standard disclaimer ---

1 0

RE: Multihoming - Resilience or Independence
by Lu, Ping 11 Oct '01

11 Oct '01

That is very common procedure. In doing that at least 2 entries will be in the global routing table. And that's exactly the intension of the customer so in case one path failed(thus withdrwan properly ) the other path will kick in. If the provider B lost both connections to the customer and provider A then it lost the FULL correct routing table then this is not the issue of filtering /20 anymore. If provider B configure their BGP sessions right then it should withdraw all the routes from provider A and the customer. Ping Lu Cable & Wireless USA Network Tools and Analysis Group W: +1-703-292-2359 E: plu(a)cw.net > -----Original Message----- > From: BLOCKI,JACEK (HP-Poland,ex1) [mailto:jacek_blocki@hp.com] > Sent: Thursday, October 11, 2001 2:12 PM > To: Lu, Ping; lir-wg(a)ripe.net; routing-wg(a)ripe.net > Subject: RE: Multihoming - Resilience or Independence > > > Hi, > The customer will of course use 2 ISP - as requested. He will > use addresses > coming from a single PA block advertised by two providers. Of > course you > need to ensure 100% up connection between border routers. If > connection > between border routers is down one of them has to stop > advertising PA block. > Otherwise you have problem with return traffic. Imagine > provider B lost > connection to both customer and provider A. If B is still > advertising common > block some ASes may decide to reach customer over B, but B > cannot reach > customer network... In order to deliver proposed service > providers A and B > have to be connected with redundant network, which is quite common. > > Vicious circle (sp) as I see it: > *Providers want small routing tables > *Customers 100% up service so they as for resilience > *Providers usually have single POP in area > *Customers go BGP creating problem for community (table grows) and > themselves (BGP expert costs) > > Solution: > Providers enter into cooperation so they can advertise > customer addresses > over 2 independent BGP speakers. This idea is not very different from > existing aggregation. Specific routing information is hidden > in a group of > ASes instead of single AS. Of course you need some > administrative effort > (e.g. consistent billing) but from technical point of view it seems > feasible. I may be wrong, it won't be for the first time ;-) > Regards, > Jacek > > > -----Original Message----- > From: Lu, Ping [mailto:PLu@cw.net] > Sent: Thursday, October 11, 2001 5:02 PM > To: 'BLOCKI,JACEK (HP-Poland,ex1)'; lir-wg(a)ripe.net; > routing-wg(a)ripe.net > Subject: RE: Multihoming - Resilience or Independence > > > > > -----Original Message----- > > From: BLOCKI,JACEK (HP-Poland,ex1) [mailto:jacek_blocki@hp.com] > > Sent: Wednesday, October 10, 2001 1:10 PM > > To: lir-wg(a)ripe.net; routing-wg(a)ripe.net > > Subject: RE: Multihoming - Resilience or Independence > > > > > > Hi, > > It seems multihoming discussion is an example of vicious > > circle (sp): More > > specific routes result in larger routing table but we want > > those routes and > > small table... Let me suggest a blasphemy: CIDR being > > advertised from two > > ASes. They say it should be only one, but why? In my > opinion there is > > nothing wrong in advertising CIDR from two ASes as long as > > you can guarantee > > each border router advertises CIDR only if it can reach it. > > Imagine the > > following construction: > > AS-A--\ > > +(OSPF) --(CIDR) > > AS-B--/ > > > > That's the problem. When customer want to be muitl-homed you > can't tell them > to use which ISP ? If the original prefix are not in the CIDR > with the other > ISP > then the customer have to change all their IP address. If > they don't then > the second > ISP have to leak the original prefix into global routing table. > > The question is how do you convince your customer to change > all their IP > addresses to > have a second link without the risk of losing its business ? > > > > Each BGP speaker advertises CIDR if and only if it learned > > about it from > > OSPF. It can be done, if you don't know how I'll forward > you a working > > example. Each border router generates a default router and > > injects it into > > OSPF. From technical point of view I see no reasons why it > > should not work. > > What you need is: > > * An agreement between ISPs > > * Change in procedure making such a union of ASes an > > officially blessed > > solution so nobody would dare to hinder cooperation with filters. > > * Optionally you may need a separate CIDR, since both ASes > > have to advertise > > same prefix. You need it if each ISP wants to have private > > customers in > > addition to shared ones. > > > > The customer has "an independent connection to two ISPs" > > which is the quest > > item. I see more commercial than technical problems with such > > a solution. > > However my expertise is limited and somebody can point > > drawbacks I cannot > > see. Feel free to burn me on a stake, that's the right way of > > treating a > > blasphemy ;-) > > Regards, > > Jacek > > > > It is a good idea but not easy to implement under the > pressure of making > revenue. > > > Ping Lu > Cable & Wireless USA > Network Tools and Analysis Group > W: +1-703-292-2359 > E: plu(a)cw.net >

1 0

RE: Multihoming - Resilience or Independence
by BLOCKI,JACEK (HP-Poland,ex1) 11 Oct '01

11 Oct '01

Hi, The customer will of course use 2 ISP - as requested. He will use addresses coming from a single PA block advertised by two providers. Of course you need to ensure 100% up connection between border routers. If connection between border routers is down one of them has to stop advertising PA block. Otherwise you have problem with return traffic. Imagine provider B lost connection to both customer and provider A. If B is still advertising common block some ASes may decide to reach customer over B, but B cannot reach customer network... In order to deliver proposed service providers A and B have to be connected with redundant network, which is quite common. Vicious circle (sp) as I see it: *Providers want small routing tables *Customers 100% up service so they as for resilience *Providers usually have single POP in area *Customers go BGP creating problem for community (table grows) and themselves (BGP expert costs) Solution: Providers enter into cooperation so they can advertise customer addresses over 2 independent BGP speakers. This idea is not very different from existing aggregation. Specific routing information is hidden in a group of ASes instead of single AS. Of course you need some administrative effort (e.g. consistent billing) but from technical point of view it seems feasible. I may be wrong, it won't be for the first time ;-) Regards, Jacek -----Original Message----- From: Lu, Ping [mailto:PLu@cw.net] Sent: Thursday, October 11, 2001 5:02 PM To: 'BLOCKI,JACEK (HP-Poland,ex1)'; lir-wg(a)ripe.net; routing-wg(a)ripe.net Subject: RE: Multihoming - Resilience or Independence > -----Original Message----- > From: BLOCKI,JACEK (HP-Poland,ex1) [mailto:jacek_blocki@hp.com] > Sent: Wednesday, October 10, 2001 1:10 PM > To: lir-wg(a)ripe.net; routing-wg(a)ripe.net > Subject: RE: Multihoming - Resilience or Independence > > > Hi, > It seems multihoming discussion is an example of vicious > circle (sp): More > specific routes result in larger routing table but we want > those routes and > small table... Let me suggest a blasphemy: CIDR being > advertised from two > ASes. They say it should be only one, but why? In my opinion there is > nothing wrong in advertising CIDR from two ASes as long as > you can guarantee > each border router advertises CIDR only if it can reach it. > Imagine the > following construction: > AS-A--\ > +(OSPF) --(CIDR) > AS-B--/ > That's the problem. When customer want to be muitl-homed you can't tell them to use which ISP ? If the original prefix are not in the CIDR with the other ISP then the customer have to change all their IP address. If they don't then the second ISP have to leak the original prefix into global routing table. The question is how do you convince your customer to change all their IP addresses to have a second link without the risk of losing its business ? > Each BGP speaker advertises CIDR if and only if it learned > about it from > OSPF. It can be done, if you don't know how I'll forward you a working > example. Each border router generates a default router and > injects it into > OSPF. From technical point of view I see no reasons why it > should not work. > What you need is: > * An agreement between ISPs > * Change in procedure making such a union of ASes an > officially blessed > solution so nobody would dare to hinder cooperation with filters. > * Optionally you may need a separate CIDR, since both ASes > have to advertise > same prefix. You need it if each ISP wants to have private > customers in > addition to shared ones. > > The customer has "an independent connection to two ISPs" > which is the quest > item. I see more commercial than technical problems with such > a solution. > However my expertise is limited and somebody can point > drawbacks I cannot > see. Feel free to burn me on a stake, that's the right way of > treating a > blasphemy ;-) > Regards, > Jacek > It is a good idea but not easy to implement under the pressure of making revenue. Ping Lu Cable & Wireless USA Network Tools and Analysis Group W: +1-703-292-2359 E: plu(a)cw.net

1 0

ANNOUNCEMENT: Public Release of Asused
by Manuel Valente 11 Oct '01

11 Oct '01

Greetings, The RIPE NCC is pleased to announce the release of asused version 3.64. Asused is a tool to check and summarise address space registered in the RIPE database. For a given list of allocations the tool prints various information concerning the allocations and the assignments they contain. The software is written in Perl and uses several Perl modules. Please grab it from: ftp://ftp.ripe.net/tools/asused-3.64.tar.gz ChangeLog for version 3.64: 3.64 Wed Oct 10 14:22:24 2001 * Fixed recognition of inetnum objects with <tab> sign in the range. * Broken dates in changed: field now handled properly, without exiting the program. * Fixed failure in case of registry with no allocations. * Recognize mixed case status: fields. 3.63 Tue Aug 21 13:52:04 2001 * Fixed netname handling according to the RIPE223 - now, case doesn't matter, so netnames with different letter-case still match. * There have been reports about Makefile failure with Perl 5.002, due changed format in the output of the MakeMaker. This is fixed now, by using our own stubs in the Makefile.PL. * Couple of strange inetnum objects were found in the DB, namely - 192.0.0.0 - 192.255.255.255, which broke proper recognition of the allocation. Now they are recognized by asused. Still, it's possible that there are other similar objects in DB. Note: Versions 3.61 and 3.62 were not publicly released. -- Manuel Valente - Software Manager - RIPE NCC

1 0

RE: Multihoming - Resilience or Independence
by Lu, Ping 11 Oct '01

11 Oct '01

> -----Original Message----- > From: BLOCKI,JACEK (HP-Poland,ex1) [mailto:jacek_blocki@hp.com] > Sent: Wednesday, October 10, 2001 1:10 PM > To: lir-wg(a)ripe.net; routing-wg(a)ripe.net > Subject: RE: Multihoming - Resilience or Independence > > > Hi, > It seems multihoming discussion is an example of vicious > circle (sp): More > specific routes result in larger routing table but we want > those routes and > small table... Let me suggest a blasphemy: CIDR being > advertised from two > ASes. They say it should be only one, but why? In my opinion there is > nothing wrong in advertising CIDR from two ASes as long as > you can guarantee > each border router advertises CIDR only if it can reach it. > Imagine the > following construction: > AS-A--\ > +(OSPF) --(CIDR) > AS-B--/ > That's the problem. When customer want to be muitl-homed you can't tell them to use which ISP ? If the original prefix are not in the CIDR with the other ISP then the customer have to change all their IP address. If they don't then the second ISP have to leak the original prefix into global routing table. The question is how do you convince your customer to change all their IP addresses to have a second link without the risk of losing its business ? > Each BGP speaker advertises CIDR if and only if it learned > about it from > OSPF. It can be done, if you don't know how I'll forward you a working > example. Each border router generates a default router and > injects it into > OSPF. From technical point of view I see no reasons why it > should not work. > What you need is: > * An agreement between ISPs > * Change in procedure making such a union of ASes an > officially blessed > solution so nobody would dare to hinder cooperation with filters. > * Optionally you may need a separate CIDR, since both ASes > have to advertise > same prefix. You need it if each ISP wants to have private > customers in > addition to shared ones. > > The customer has "an independent connection to two ISPs" > which is the quest > item. I see more commercial than technical problems with such > a solution. > However my expertise is limited and somebody can point > drawbacks I cannot > see. Feel free to burn me on a stake, that's the right way of > treating a > blasphemy ;-) > Regards, > Jacek > It is a good idea but not easy to implement under the pressure of making revenue. Ping Lu Cable & Wireless USA Network Tools and Analysis Group W: +1-703-292-2359 E: plu(a)cw.net

1 0

RE: Multihoming - Resilience or Independence
by Boyan Krosnov 11 Oct '01

11 Oct '01

> Your prefixes (let's say /24) has to be announced by two ISPs > so the packets > will come back to you. If ISP A's link failed then that /24 > with ISP A in > the > ASPATH will be withdrawn. And ISP B's /24 will become active. > So in the > global > routing table there will be TWO /24 with different ASPATH. > > If this /24 is in ISP A's PA space then it saves one but the > one from ISP > B's one > NEEDs to be in the global routing table. Either I didn't understand you right, or you were wrong. It does not save a route in the DFZ. Let's say the customer gets A/24 out of proider A's A/17 prefix. B has to announce A/24 to the internet along with it's current prefixes. A has to announce A/17 AND A/24 to the internet. If A advertises A/17 only all the traffic for A/24 will go through B (because of the longest match rule), and I think that's not what that customer wants. So Being multihomed to two ASs with NAT or two addreses on each host does not create an extra route in the DFZ. Being multihomed to a single AS does not create an extra route in the DFZ. Being multihomed to two ASs (It doesn't matter PA or PI, separate AS or anounced by the two ASs. Prefix out of PA, separate AS is generally recognised to be the better technical solution.) does create at least 1 extra prefix and 2 extra as-paths in the DFZ. These are the facts about growth of the DFZ routing table caused by multihoming as I see them. There are of course other reasons for the growth like providers advertising 14 separate /24s when they can agregate it to /20 because "We were trying to do some input channel load balacing and have left this anouncements for months because our administrator does not have the time to deal with this thing, he/she has more important work to do". Now it's time to think how we can make people prefer the first 2 possibilities over the third. We've had enough of technical discussion I think. Regards, Boyan Krosnov, CCNP Network Administrator Lirex Net phone: +359-2-91815

1 0

changed attribute proposal and not only
by Andrei Robachevsky 11 Oct '01

11 Oct '01

Dear Colleagues, Please find enclosed the proposal of modification of the changed attribute that was presented at the RIPE-40 meeting in Prague. However, another proposal (or rather idea) came up in light of recent discussions regarding timestamping and cleanup procedures in the RIPE Database. Let me present you this idea first. As was pointed by the two proposals there is a need to store and display status information about an object in the database, such as last update time, number of references, etc. At the same time in order to be reliable this information should be automatically generated by the database software and be outside users' control. Regarding "changed attribute" proposal people felt that this changes some of the basic principles of the RIPE Database. Recent Ulf's proposal makes things even worse since it requires that modification to an object may be caused by an update of another object. But in fact these attributes/facilities do not represent an object, but rather status information about the object. Extending Ulf's analogy with the Unix FS, they are part of the inode, not the data blocks. So status information (object metadata) may be stored and displayed separately from the object itself by, say, using a flag. Much like we display the contents using cat, but status using ls. And this information will be used for cleanup procedure, reliable timestamping, etc. This will not affect any object template, preserve users' data and add flexibility to the database. Something like this: $ whois -q meta RC1234-RIPE ... person: Referenced Contact nic-hdl: RC1234-RIPE ... changed: refcon(a)somedom.dom 20011005 source: RIPE %metainfo: RC1234-RIPE %update-time: 2001-10-05 16:50:07 %ref-time: 2001-10-06 09:17:33 %ref-count: 3 Regards, Andrei Robachevsky DB Group Manager RIPE NCC

1 0

40.5: RFC 2050 mailinglist
by Hans Petter Holen 10 Oct '01

10 Oct '01

Dear lir-wg, At the meeting in Praha we briefly discussed how to contribute to the global effort to revise RFC 2050. We were informed that ARIN had set up a working group for this, with a mailing list. It was suggested that we might join the mailing list hosted by ARIN to participate in this work, thus forming a truly global effort on the matter. Ray Plzak the President of ARIN invited the RIPE community to do so. After positive reactions from the participants, and no voices against, the chair declared consensus on this matter. The details of the list is: * List Address: 2050-wg(a)arin.net * Subscription is open to all interested parties * List is archived More information about the list is available at: http://www.arin.net/members/mailing.htm The charter from ARINs announcement: "The objective of the RFC 2050 Working Group is to address the issues relating to relevance of RFC 2050 to the needs of today's Internet registry system. The group will evaluate RFC 2050 and propose a method of replacing it with a new document or documents. Once consensus has emerged on the process that will be used to replace RFC 2050, the working group will cooperatively develop its replacement. The working group will work in coordination with the other Regional Internet Registries, who will conduct a similar review process in their respective regions. " Hans Petter Holen WG Chair ARIN announcement: http://www.arin.net/mailinglists/arin-announce/0121.html From: Member Services (memsvcs(a)arin.net) Date: Mon Sep 10 2001 - 11:24:00 EDT ---------------------------------------------------------------------------- ---- At the last ARIN Open Policy meeting in San Francisco concerns were raised about RFC 2050. It was noted that this document has aged and that certain portions of it no longer reflected the current state of RIR policy and operations. Similar concerns have been raised in the APNIC and RIPE NCC regions. Accordingly, ARIN is establishing a working group to further examine this issue. Mark McFadden has volunteered to chair the 2050WG. The charter for the working group follows: The objective of the RFC 2050 Working Group is to address the issues relating to relevance of RFC 2050 to the needs of today's Internet registry system. The group will evaluate RFC 2050 and propose a method of replacing it with a new document or documents. Once consensus has emerged on the process that will be used to replace RFC 2050, the working group will cooperatively develop its replacement. The working group will work in coordination with the other Regional Internet Registries, who will conduct a similar review process in their respective regions. An email list for the 2050WG has been established. Subscription information can be found at the following URL: http://www.arin.net/members/mailing.htm All interested individuals are encouraged to participate in this discussion. This will be an agenda item at the next ARIN Open Policy meeting in Miami. Ray Plzak President

1 0

RE: Multihoming - Resilience or Independence
by BLOCKI,JACEK (HP-Poland,ex1) 10 Oct '01

10 Oct '01

Hi, It seems multihoming discussion is an example of vicious circle (sp): More specific routes result in larger routing table but we want those routes and small table... Let me suggest a blasphemy: CIDR being advertised from two ASes. They say it should be only one, but why? In my opinion there is nothing wrong in advertising CIDR from two ASes as long as you can guarantee each border router advertises CIDR only if it can reach it. Imagine the following construction: AS-A--\ +(OSPF) --(CIDR) AS-B--/ Each BGP speaker advertises CIDR if and only if it learned about it from OSPF. It can be done, if you don't know how I'll forward you a working example. Each border router generates a default router and injects it into OSPF. From technical point of view I see no reasons why it should not work. What you need is: * An agreement between ISPs * Change in procedure making such a union of ASes an officially blessed solution so nobody would dare to hinder cooperation with filters. * Optionally you may need a separate CIDR, since both ASes have to advertise same prefix. You need it if each ISP wants to have private customers in addition to shared ones. The customer has "an independent connection to two ISPs" which is the quest item. I see more commercial than technical problems with such a solution. However my expertise is limited and somebody can point drawbacks I cannot see. Feel free to burn me on a stake, that's the right way of treating a blasphemy ;-) Regards, Jacek

1 0

RE: Multihoming - Resilience or Independence
by Lu, Ping 10 Oct '01

10 Oct '01

> -----Original Message----- > From: Steven Bakker [mailto:steven@icoe.att.com] > Sent: Wednesday, October 10, 2001 11:08 AM > To: Nipper, Arnold > Cc: Stephen Burley; Gert Doering; Dave Pratt; lir-wg(a)ripe.net; > routing-wg(a)ripe.net > Subject: Re: Multihoming - Resilience or Independence > > > On Wed, 2001-10-10 at 15:24, Nipper, Arnold wrote: > > Of course another prefix and another AS is added to the > routing table > > but thereby you are witdrawing at least more than two (both > prefs as well as > > AS). ==> table gets smaller. > > Your math is confusing... or maybe I'm misunderstanding the obvious. > Yes, multihoming adds another AS and (at least one) (longish) > prefix to > the routing table. But in your reasoning, which two prefixes > and one AS > are being withdrawn? If I resiliently multi-home to one ISP, > and I get > my address space from its block, I do not add any prefix or AS to the > (global) routing table. Even if I multi-home to two ISPs and > selectively NAT depending on the outgoing connection, I'll be NAT-ing > into the respective ISP's address space (which I presume is properly > aggregated), adding no prefix or AS. Your prefixes (let's say /24) has to be announced by two ISPs so the packets will come back to you. If ISP A's link failed then that /24 with ISP A in the ASPATH will be withdrawn. And ISP B's /24 will become active. So in the global routing table there will be TWO /24 with different ASPATH. If this /24 is in ISP A's PA space then it saves one but the one from ISP B's one NEEDs to be in the global routing table. The worse case: ISP A only announces aggregate block even if the link failed then the traffic will still go thru ISP A then get dropped. Ping Lu Cable & Wireless USA Network Tools and Analysis Group W: +1-703-292-2359 E: plu(a)cw.net

1 0