- NZNOG - lists.nznog.org

Fwd: Asia Pacific Peering Guidebook Reviewer Request
by Joe Abley 16 Jun '04

16 Jun '04

Begin forwarded message: > From: "William B. Norton" <wbn(a)equinix.com> > Date: 16 June 2004 12:49:01 EDT > To: nanog(a)merit.edu > Cc: wbn(a)equinix.com > Subject: Asia Pacific Peering Guidebook Reviewer Request > > > Hi all - > > If you have no interest in Internet Peering in Asia, read no further. > Otherwise,... > > Over the last year I have been working with the Peering Coordinator > Community, particularly those that have built into and throughout > Asia, to document what is *different* about peering in Asia as > compared to say the U.S. and Europe. In February of this year I > pulled together a group of Peering Coordinators for the APRICOT > Peering Track. I asked them to share Peering processes that they found > different, interesting, counter intuitive, or unexpected. This white > paper documents the lessons learned and insights gained from Peering > Coordinators experienced with peering in Asia. > > The Asia Pacific Peering Guidebook follows the path set by "The > Evolution of the US Peering Ecosystem" white paper that I presented at > the last NANOG, but focuses on the common dynamics and experiences > that folks shared with me over the last year and at the APRICOT > Peering Track. First, the paper documents what I call the "Foreign > Tier 1 Peering Dynamic" that casts a Tier 1 ISP in one Peering > Ecosystem as a Tier 2 ISP in another. We focus on the motivations, the > reasons why this occurs as ISPs expand globally. > > Next, Peering Coordinators shared five reasons that they expanded > into/within Asia: > 1) Incumbent Tier 1 ISPs to peer their routes outside their home market > 2) Meet US Tier 1 ISP Peering Prerequisites > 3) Customers want them in Asia > 4) Global Marketing Benefits > 5) Sell transit in a more attractive transit market > > using four methods of interconnection: > 1) Extend into foreign country > 2) reciprically peer in each others coountry > 3) Half-Circuit peering > 4) Buy an ISP with Tier 1 peering in the foreign ecosystem > > The Peering Coordinators shared nine lessons, things they wish they > had known before expanding into Asia: > 1) Tier 1 ISPs will not want to peer in their home market > 2) There are several challenges peering in Asia > 2a) Peering Process is different - peering@ does not uniformly work > 2b) Many language zones > 2c) Many time zones - logisitics issues here > 2d) Oceans add cost, latency - transpacific transport prices estimated > in the paper > 2e) local loops still a major expense > 2f) transit prices highly variable across countries in Asia - transit > street prices quoted > 3) Creative peering deals > 4) Watch for Inadvertent Tromboning > 5) Local Presence is required > 6) Separation of Int'l Transit from Domestic Transit in some markets > 7) Content that transcends the language barrier is not allowed to be > hosted in country > 8) No true regional content in Asia > 9) Content Peering works in Asia > 10) Understand the Internet Peering Ecosystem prior to building in > > On this last point, the paper enumerates several data points including > naming the Tier 1's, the emerging broadband players, the peering > inclinations, etc. for Hong Kong, Tokyo, Sydney and Singapore. Some of > the Yahoo!Broadband activities for example in Japan show that some > places in Asia are leap-frogging the US in terms of next-gen access > bandwidth. How would you like 40Mbps DSL access for $50/mo or FTTH for > $100/mo? > > I'm at the point where I'd like to do a dozen or so final paper > walk-throughs, where I email the white paper and walk interested folks > through the paper to solicit comments and suggestions. I've found > that talking people through the paper is this is the most effective > way for me to make sure the paper (and subsequent presentation) flow > well, and also increases the likelihood of me getting some feedback. > > If you have any interest in the topic and can spare about 30 minutes > for a walkthrough over the next few weeks, send me a note and let's > schedule a time I can talk you through the paper. I'm also planning a > group walk through (a concall basically) towards the end of the month. > > As with all the Peering White Papers, this one will be made freely > available to the community when it is finished. > > Bill > > Here are the other Peering White Papers that are available (google for > "William B. Norton" to find them on the web, or email wbn(a)equinix.com > with the white paper title in the subject line and I'll send you the > latest version of the white paper.) > > 1. Interconnection Strategies for ISPs documents two dominant > methods ISPs use to interconnect their networks. Over 200 ISPs helped > create this white paper to identify when Internet Exchange Points make > sense and the Direct Circuit interconnect method makes sense. > Financial Models included in the paper quantify the tradeoffs between > these two methods. All relevant data points are footnoted as to > source. > 2. Internet Service Providers and Peering answers the questions: > "What is Peering and Transit? What are the motivations for Peering? > What is the ISP Peering Coordinators Process for obtaining peering? > What are criteria for IX selection?" > 3. A Business Case for Peering builds upon the previous white > papers but focuses on the questions important to the Chief Financial > Officer: "When does Peering make sense from a financial standpoint? > When do all the costs of Peering get completely offset by the cost > savings?" > 4. The Art of Peering: The Peering Playbook builds on the > previous white papers by asking the Peering Coordinators to share the > "Tricks of the Trade", methods of getting peering where otherwise they > might not be able to get peering. These 20 tactics range from the > straight forward to the obscure, from the clever to the borderline > unethical. Nonetheless, Peering Coordinators might be interested in > field-proven effective ways of obtaining peering in this highly > controversial white paper. > 5. The Peering Simulation Game finishes up my half day Peering > Tutorial by engaging the audience in the role of the Peering > Coordinator. Each ISP in turn rolls the dice, expands their network, > collects revenue for each square of customer traffic, and pays transit > fees to their upstream ISP. They quickly learn that if they peer with > each other, the costs of traffic exchange are much less, but they need > to negotiate how to cover the costs of the interconnect. ISP Peering > coordinators have commented on how close the peering simulation game > is to reality in terms of the dialog that takes place. > 6. Do ATM-based Internet Exchange Points Make Sense Anymore? > Applies the "Business Case for Peering" financial models to ATM and > Ethernet-based IXes using current market prices for transit, > transport, and IX Peering Costs. > 7. The Evolution of the U.S. Peering Ecosystem, introduces and > focuses on several fundamental changes in the Peering Ecosystem > spurred by several events following the telecom collapse of 1999/2000. > 8. The Art of Peering: The IX Playbook follows the same tact as > The Peering Playbook; we first introduce the framework theory of how > and why IXes are valuable from an economic perspective. We then > enumerate about a dozen tactics IXes use to get over the "Start Up > Hump", to build a strong critical mass of participants, and finally, > defense tactics to maintain that population. (To be released at a > future date.) >

1 0

CallPlus prefix list update
by Steve Hawken 16 Jun '04

16 Jun '04

Hi, For networks peering with CallPlus AS 9790 please update your prefix list as below: ip prefix-list callplus-networks description callplus networks contact noc(a)callplus.co.nz ip prefix-list callplus-networks seq 5 permit 202.0.151.0/24 ip prefix-list callplus-networks seq 10 permit 202.36.244.0/23 le 24 ip prefix-list callplus-networks seq 15 permit 202.37.129.0/24 ip prefix-list callplus-networks seq 20 permit 202.49.34.0/23 le 24 ip prefix-list callplus-networks seq 25 permit 202.49.62.0/24 ip prefix-list callplus-networks seq 30 permit 202.50.109.0/24 ip prefix-list callplus-networks seq 35 permit 202.50.170.0/24 ip prefix-list callplus-networks seq 40 permit 202.180.64.0/18 le 24 ip prefix-list callplus-networks seq 45 permit 203.97.152.0/24 ip prefix-list callplus-networks seq 50 permit 203.100.223.208/20 le 24 ip prefix-list callplus-networks seq 55 permit 203.184.0.0/18 le 24 ip prefix-list callplus-networks seq 60 permit 203.215.142.0/23 le 24 ip prefix-list callplus-networks seq 65 permit 210.55.111.0/24 ip prefix-list callplus-networks seq 70 permit 210.55.254.0/24 Cheers, Steve Steve Hawken Senior Network Engineer DDI ++ 64 9 919 6038 Mobile ++ 64 21 778 891 Email steveh(a)callplus.co.nz <mailto:steveh(a)callplus.co.nz>

1 0

NZNOG 2005 in Hamilton
by Donald Neal 15 Jun '04

15 Jun '04

Planning is now under way for the next NZ Network Operators' Group conference. This is to be hosted by the WAND group (Thanks again, guys!) at The University of Waikato's main campus in Hamilton in the first week of February 2005. If you have an opinion on the form this should take, please speak up now. If not, you're likely to end up with something bearing a strong resemblance to this year's gathering, with tutorials on Wednesday and a stream of presentations on Thursday and Friday. (That's 2-4 February.) Don't rush to volunteer to present just yet - we should have a proper Call for Presenters out around the end of this month - but now is the time for people who are likely to attend to suggest topics they'd like to know more about, people they'd like to hear from, or changes to the way things have been done in the past. Please post suggestions either to me or to the list as seems appropriate. I'll try to provide a digest of responses by next Friday, 25 June. - Donald Neal Donald Neal | Bubble and squeak Technical Specialist | - Boiled mouse. Operations Engineering | [Martin Neal] Integration & Services Division +----------------------- Alcatel NZ Ltd - Telecom's network operations manager ------------------------------------------------------------------------------ "This communication, including any attachments, is confidential. If you are not the intended recipient, you should not read it - please contact me immediately, destroy it, and do not copy or use any part of this communication or disclose anything about it. Thank you. Please note that this communication does not designate an information system for the purposes of the Electronic Transactions Act 2002." ------------------------------------------------------------------------------

1 0

Re: Dynamic address space in SORBS causes mail rejection
by Richard Cox 15 Jun '04

15 Jun '04

On Wed, 16 Jun 2004 11:14:12 +1200 "Craig Box" <craig(a)itpartners.co.nz> wrote: | While we're on the subject of mail rejections, I look after a couple | of sites on a TelstraClear Tempest connections, both with IP addresses | on 203.97.48.0/21. Unfortunately, spam DNS blacklist SORBS list them | as being on Dynamic IP space: | | 203.97.48.x found in Dynamic IP Space (Cable, DSL & Dial Ups) | Address or Block: 203.97.48.0 / 21 | Information: Dynamic IP address, use your ISPs mail server | | ..and as such a number of sites (a number growing by about one a day, | just the ones I get told about) are blocking mail based on the SORBS | listing. The point being that the inbound mail administrators have the right to decide whether to accept connections from hosts in dynamic-IP ranges. | Some months ago I got in touch with TC's faults desk and they | claim to have emailled off to SORBS to have this reclassified. | | Nothing changed, so I got in touch with our local tech contact at TC; | he said he'd fire it off to the ISP guys but I've not heard back from | him either. All true. On Jan 30 The Telstra Abuse Desk wrote to SORBS about the block being labelled as "dynamic" and asked for "a copy of the headers of the spam". SORBS replied, explaining that the DUHL list identified their IP's DHCP status, not that any spam had been sent. Telstra Abuse Desk didn't respond to that reply. On March 11 Mark Seward wrote in again from Telstra, asking for the block status to be changed: SORBS replied, explaining that for security reasons such a request would need to come from the designated APNIC PoC for the IP block (which currently would be netobjs(a)clear.net.nz) However that reply got an auto-response saying that Mark Seward was no longer with Telstra, and mail should now go to Peter Ambrose and/or Sid Jones, so SORBS dutifully forwarded the previous correspondence to those two people, hoping they'd get the appropriate person to write in to authorise the change. But nothing further was heard until April 7 when Jason, as "isp.duty(a)team.telstraclear.co.nz" wrote in asking how to get the block's status changed, and he was sent a similar answer to that which had gone to Mark Seward, Peter Ambrose, and Sid Jones: but still nothing was heard from the designated PoC. | I've asked around about this and the best suggestions are "you're an | ISP customer, you should be smarthosting through an ISP mail server." | Problems with this include the fact that we no longer have any control | over retry/routing/any of the niceties of controlling your own outbound | SMTP; the fact that clear.net.nz mailservers are currently being eaten | by Zafi and mail delivery will probably slow down; and the fact that | we "just plain shouldn't have to." If you're on static IP, then as you say, you shouldn't have to. TelstraClear should have been able to follow the simple instructions to get this changed, normally a very straightforward process! | I'm not sure where to go from here. It's possible that if I ask TC | to set the PTR for each affected customer to smtp.$CLIENT and then | manually request that it's removed from SORBS, it may be done. If an IP is to send mail, that is a good suggestion in its own right. DNS should also have a sensible TTL - at least 12 hours is reasonable. -- Richard Cox

1 0

Akamai outage
by Juha Saarinen 15 Jun '04

15 Jun '04

http://www.infoworld.com/article/04/06/15/HNakamaioutage_1.html Not sure if NZ is affected too, but have noticed that Windows Update is unusually slow. -- Juha

2 1

Dynamic address space in SORBS causes mail rejection
by Craig Box 15 Jun '04

15 Jun '04

Hi everyone. While we're on the subject of mail rejections, I look after a couple of sites on a TelstraClear Tempest connections, both with IP addresses on 203.97.48.0/21. Unfortunately, spam DNS blacklist SORBS (http://www.dnsbl.au.sorbs.net/) list them as being on Dynamic IP space: 203.97.48.x found in Dynamic IP Space (Cable, DSL & Dial Ups) Address or Block: 203.97.48.0 / 21 Information: Dynamic IP address, use your ISPs mail server ...and as such a number of sites (a number growing by about one a day, just the ones I get told about) are blocking mail based on the SORBS listing. Some months ago I got in touch with TC's faults desk and they claim to have emailled off to SORBS to have this reclassified. Nothing changed, so I got in touch with our local tech contact at TC; he said he'd fire it off to the ISP guys but I've not heard back from him either. I've asked around about this and the best suggestions are "you're an ISP customer, you should be smarthosting through an ISP mail server." Problems with this include the fact that we no longer have any control over retry/routing/any of the niceties of controlling your own outbound SMTP; the fact that clear.net.nz mailservers are currently being eaten by Zafi and mail delivery will probably slow down; and the fact that we "just plain shouldn't have to." We have a hosted mail server of our own that we could use as a smarthost, however it will be a significant increase in load on this machine where we have a perfectly good mail server sitting at the clients site. I agree completely about customers not running mail service on dial-up lines, but a Tempest connection with a static IP address is a business service, not a 56k modem with an infected Windows box behind it. I'm not sure where to go from here. It's possible that if I ask TC to set the PTR for each affected customer to smtp.$CLIENT and then manually request that it's removed from SORBS, it may be done. Their FAQ (http://www.dnsbl.au.sorbs.net/DUL-FAQ.html) suggests I'm more likely to be asked to have my ISP ask. I'd like to see TC deal to this, perhaps by putting more information in the WHOIS db about what space is static and what is dynamic, and possibly changing PTR records appropriately. According to the DNSBL faq, the contact listed in the RIR can have it changed - it would be really nice if this problem could go away. If there are any other suggestions that anyone has about how to deal with this problem, please let me know. Regards, Craig Box Phone 07 957 2653 IT Partners Ltd Fax 07 957 2659 PO Box 9361 Mobile 021 475 869 Hamilton, New Zealand _________________________________________________________________ #include <standard-disclaimer.h> _________________________________________________________________

1 0

Map/schematic over NZ Internet connectivity?
by Juha Saarinen 15 Jun '04

15 Jun '04

Has the NZ Internet connectivity been mapped out yet? Maybe I'm not googling right, but I can't find any resource which lists national and international connectivity for New Zealand. -- Juha

6 9

RE: [nznog] TelstraClear Inhternational / National Traffic billing
by Donald Neal 15 Jun '04

15 Jun '04

> On Tue, 15 Jun 2004, Joe Abley wrote: > > > On 15 Jun 2004, at 08:22, Drew Broadley wrote: > > > > > The end is nigh. > > > > The doomsaying is all very dramatic, but where's the reasoning? > > > > If it suddenly costs money to talk to networks which are > customers of > > TCL, it surely makes *more* sense to peer with them > directly, not less. Indeed. Peering by US content providers is one of the big changes identified in http://www.nanog.org/mtg-0405/pdf/norton.pdf . Perhaps sadly, one of the possible futures for the US identified by the author in answer to questions about his presentation at the most recent NANOG was the New Zealand model, in which charges aren't made for national traffic. - Donald Neal Donald Neal | Bubble and squeak Technical Specialist | - Boiled mouse. Operations Engineering | [Martin Neal] Integration & Services Division +----------------------- Alcatel NZ Ltd - Telecom's network operations manager ------------------------------------------------------------------------------ "This communication, including any attachments, is confidential. If you are not the intended recipient, you should not read it - please contact me immediately, destroy it, and do not copy or use any part of this communication or disclose anything about it. Thank you. Please note that this communication does not designate an information system for the purposes of the Electronic Transactions Act 2002." ------------------------------------------------------------------------------

1 0

TelstraClear Inhternational / National Traffic billing
by neil gardner 15 Jun '04

15 Jun '04

Hi guys - I just spent 5 minutes talking to Telstra in between 45 minutes of working while on hold... The question I asked was: If TelstraClear go ahead with cutting the links to a couple of big national network access points (I was speaking to a non-tech), what will happen to the national / international traffic billing. This person had zero clue but came back ten minutes later and stated she'd been talking to someone that knew exactly what I was talking about. His statement was that "Shortly the International / National traffic distinction will be phased out and there will be only one class of traffic. The cost will be similar to the current International rate. But don't worry since it only matters if you go over your limit" (The lady relayed his comment and that's as close to a direct quote as I can recall) Does this mean that TelstraClear are going to adopt the Telecom Jetstream traffic charging model? Anyone in the know care to correct me if I was given false information by junior Telstra Staff? Or, horribly, was this statement correct? (I specifically wanted to know what would happen to my national / international charging if TelstraClear cut their peering, and it appears they do in fact have a plan to make sure the consumer experiences no confusion over what's National or International traffic) Cheers - Neil G NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

11 14

Re: [nznog] SMTP Servers and Reverse DNS Lookup
by Juha Saarinen 14 Jun '04

14 Jun '04

neil gardner wrote: > Um, no false positives? doesn't that mean you manually inspect all your > logs of dropped connections? Doesn't, by extension, that mean you might > as well let all the stuff through and check them at the client end? :-) Yes, I do check the logs, which is a lot easier than it may appear. No, it doesn't mean I might as well let the stuff through... -- Juha

1 0

Re: [nznog] SMTP Servers and Reverse DNS Lookup
by neil gardner 14 Jun '04

14 Jun '04

Um, no false positives? doesn't that mean you manually inspect all your logs of dropped connections? Doesn't, by extension, that mean you might as well let all the stuff through and check them at the client end? :-) Cheers - N >>> Juha Saarinen <juha(a)saarinen.org> 15/06/2004 3:30:14 p.m. >>> neil gardner wrote: > http://www.cs.niu.edu/~rickert/cf/bad-ehlo.html > > Decent understandable writeup of the "You should send the right info" > and "Don't drop email if you get the wrong info" arguments... ... and having said all that, I admit to dropping some some messages based on HELOs. Mainly viruses and spamware connections, with no false positives yet. Touch wood. -- Juha NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

2 1

Re: [nznog] SMTP Servers and Reverse DNS Lookup
by neil gardner 14 Jun '04

14 Jun '04

http://www.cs.niu.edu/~rickert/cf/bad-ehlo.html Decent understandable writeup of the "You should send the right info" and "Don't drop email if you get the wrong info" arguments... Cheers - N. >>> Juha Saarinen <juha(a)saarinen.org> 15/06/2004 2:56:14 p.m. >>> Regan Murphy wrote: > Is this sort of configuration becoming standard for organistions? Some admins think it's BOFH to do stuff like that, in the mistaken belief that it's RFC-compliant. It's a bit like fundamentalists interpreting various religious scriptures, but if they want to drop legitimate email, their choice I say. -- Juha _______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

2 1

SMTP Servers and Reverse DNS Lookup
by Regan Murphy 14 Jun '04

14 Jun '04

I was having problems sending mail to an organisation and eventually worked out that the host name and reverse dns entry did not match and therefore mail transfer was refused: ------- 220 mail.xxxbar.co.nz ESMTP NetIQ MailMarshal (v5.5.5.8) Ready ehlo hellmouth.oaxxxsystems.co.nz 554 No SMTP service here ------- As soon as I corrected my server to use the correct host name to match the reverse (vampire.oaxxxsystems.co.nz) then mail was allowed to go through. I would assume this was a recent change by the receiving organisation as I have not had problems sending them emails before. Is this sort of configuration becoming standard for organistions? I would expect that may create problems for smaller organisations sending email directly from their in-house hosted mail server for which they have no control over reverse-dns entries. -- Regards, Regan

5 5

RE: [nznog] TelstraClear Inhternational / National Traffic billing
by Craig Spiers 14 Jun '04

14 Jun '04

Well.. You could get a link to both, but the only real benefit would be splitting the cost between two (probably slightly more expensive this way as a separate circuit is needed for both - generally cheaper to upgrade a 10meg to 100meg than it is to install 2x 10meg eg. one from each telco) However, you would have a shorter data path.. As with two circuits, the data goes direct to both carriers.. With just one circuit, data to clear, goes via telecom.. (provided they keep the peering between them). -----Original Message----- From: jfp [mailto:jfp(a)clearfield.com] Sent: Tuesday, 15 June 2004 12:16 p.m. To: 'Craig Spiers' Subject: RE: [nznog] TelstraClear Inhternational / National Traffic billing > They shouldn't be connected to the internet at all! They should be > shot in the foot for having such poor routing :) > > NZ > US > NZ :) Not being in the ISP business, but a client of ISP's, could you explain that one to me. I thought the whole point of peering at the APE and WIX was that you did not need to get point to point link to other providers. Why do you NEED to get a link to one of the telco's? jfp. ------------------------------------------------------------------------ Jean-Francois Pirus <jfp(a)clearfield.com> Clearfield Software Ltd Phone (+64-9) 358 2081 4th Floor 8-10 Whitaker Place Fax (+64-9) 358 2083 P O Box 2348 Auckland, New Zealand ------------------------------------------------------------------------

1 0

Re: Map/schematic over NZ Internet connectivity?
by Kerry Milestone 13 Jun '04

13 Jun '04

The National Library of New Zealand has been given a lot! of money, i expect they will have a good map soon......... http://www.beehive.govt.nz/ViewDocument.cfm?DocumentID=19882 And an older National Business Review article: http://www.nbr.co.nz/home/column_article.asp?id=1978&cid=1&cname=Media Maybe some of it can be spent on peering networks together to help with the flow of _all_ websites ending in .nz to be harvested daily by the NatLib... This task is now part of the National Library Bill and can be thought of being the same as the Legal Deposit legislation.

1 0

xtra's MTA again??
by Dan Clark 13 Jun '04

13 Jun '04

Hi Guys, just wondering if anyone else isn't getting any mail through to XTRA. If you are and your mta is MailMarshal please let me know also, from last time there appeared to be an issue with MailMarshal threads. Kind Regards Dan Clark Network Manager Synaptic Ltd

4 3

RE: Map/schematic over NZ Internet connectivity?
by Brendan Murray 12 Jun '04

12 Jun '04

> Message: 3 > Date: Sun, 13 Jun 2004 11:26:20 +1200 > From: Juha Saarinen <juha(a)saarinen.org> > Subject: Re: [nznog] Map/schematic over NZ Internet connectivity? > To: jon.brewer(a)worldnet.att.net > Cc: 'NZ NOG' <nznog(a)list.waikato.ac.nz> > Message-ID: <40CB911C.3010600(a)saarinen.org> > Content-Type: text/plain; charset=us-ascii; format=flowed > > Jonathan Brewer wrote: > > > I tried InternetNZ with this very question about a year ago, and the > > answer was no. > > I guess there was no follow-up along the lines of "... but we're working > on it"? Give me good/adequate/useful/whatever reasons for doing it and I'll take it to the INZ tech committee. Always nice to add things to the ToDo list. More usefully - others on this list probably already have some/all of a map - where can we access that information if INZ does decide its worth putting together? And how do we keep it up to date?

2 1

Re: [nznog] Zombies
by Ewen McNeill 10 Jun '04

10 Jun '04

In message <20040610051613.GJ20881(a)stateless>, Nicholas Lee writes: >On Thu, Jun 10, 2004 at 04:58:31PM +1200, Ewen McNeill wrote: >> [... suggesting] anti-spam legislation, a legal duty be placed on people >> not to connect/allow to remain connected an insecure/0wned/infected >> system under their control. > >Unlike vehicles, access to the internet for most people is not via a >"public road". [...] >A WOF/license for driving the internet just seems like needless legisation. I actually didn't suggest a WOF/license for driving on the Internet (this time -- although I do actually think it's a good idea). What I suggested was that those who have allowed systems they own to become unsafe/damaged so that they cause harm to others should be made responsible for that harm. And it doesn't seem like needless legislation to me. It's not _my_ systems (or any of the ones I'm responsible for administering) which are sending thousands of spam messages, viruses, worms, to machines across the world. I just get to receive thousands of them a day via all these machines of people who are not taking responisbility for the systems they're connecting to a common resource (viz, the Internet). As Juha says there is a problem with patches not being available. I'd be quite happy to allow "no vendor patch" as a complete defense for the end user (although if the issue is well known the vendor should perhaps be culpable). I'd even be willing to conceed "vendor patch too recently released, couldn't patch everything" with in a reasonable amount of time of the patch coming out. Simply persuading someone to take responsibility for the never-been-patched, installed-stuff-at-random, insecure-OS-design boxes would dramatically reduce the problem. >ISPs already have the power to regulate users this way via their TOS. Its >clear though that all ISPs would have to subscribe to the above for it >to have long-term effect. The days of ISPs taking responsibility for the actions of their users appear to be a fading memory. It's unlikely any ISP would block users with infected machines these days unless all ISPs did it; the only way that all ISPs are likely to do it is if it's a legal requirement. (Otherwise it's a prisoner's dilemma situation.) Ewen

7 6

Re: [nznog] Zombies
by neil gardner 10 Jun '04

10 Jun '04

OK, serious thought here... Bear with me... I may be a) Describing an existing system or b) way off base. Transparent SMTP proxy intercepts all messages and maintains a running total of emails sent per source IP. This running total is actually stored as a time distribution (ie. 10 emails in 10 minutes, then none in 10, then 100 in ten etc) Rulebase periodically examines this time distribution for each source IP. Something like If last 3 matches are >10, >10, >10 then block for 30 minutes per strikeout(sourceIP) and increment strikeout(sourceIP) by one. After expiry of downtime, open them up again. After (say) 90 minutes of non-abhorrent behaviour, decrement strikeout(sourceip) by one. If strikeout(sourceIP) gets to threshold (Say 10) then email user and 48 hours later disable account unless they contact helpdesk and claim to have addressed the situation. You probably want cleverer rules to identify bursts of email, and possibly a training period with rules referrent to earlier data from the same source - this should be able to reliably identify legitimate high volume users - especially if you build in a time of day factor. This should quickly throttle zombies and disable the account if not fixed, but allow for brief infections to be throttled, and if fixed, NOT need manual intervention to unblock. It may impact some legit heavy email users, but mail will bounce, and it should fix itself. Just a thought. Cheers - N. >>> Simon Lyall <simon(a)darkmere.gen.nz> 10/06/2004 1:46:38 p.m. >>> It looks like there are lot of NZ based Zombies sending very large amounts of Spam ( much of it in German) to some ISPs. This has really been going hard over the last hour. I'm seeing adsl accounts at xtra, iprolink and ihug, xtra dialups etc etc. As a related issue, how do people feel about a whitelist RBL of NZ mail servers, It looks like it's not good to just whitelist NZ IPs -- Simon J. Lyall. | Very Busy | Mail: simon(a)darkmere.gen.nz "To stay awake all night adds a day to your life" - Stilgar | eMT. _______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

6 7

Re: [nznog] Zombies
by Ewen McNeill 09 Jun '04

09 Jun '04

In message <s0c876cc.089(a)LEO.ALLIEDTELESYN.CO.NZ>, "neil gardner" writes: >To be fair though, in mentioning an opt-out system I was thinking of >clueful users deliberately running their own servers that would notice >within minutes that port 25 had been blocked - not the SMTP AUTH crowd >which may be significantly less clued up and not notice why their >previously working 'Email Thingy" isn't working anymore. The clueful users can always build tunnels and route stuff through the tunnels. I do this from my laptop when working in various organisations where I still want to send mail via my own mail server (because, eg, it's the only one listed in the SPF rules). SMTP AUTH is harder, but at least as a first cut simply permitting connections that did SMTP AUTH would probably select for the "good" connections and ignore the bad ones. Much the same for STARTTLS -- anything using that is probably not malware, at least at present. Most malware barely manages to interoperate with SMTP servers let alone actually supporting "fancy" features. As I've suggested before blocking some of these services by default, and providing an "enable this service again" automatic system for clueful users would be nearly as useful as trying to do layer-7 filtering on the protocol. It could even require turning on again with every reconnection, totally doing away with the need for the ISP to store anything associating permissions with the user. (Anyone who can't automate a GET of, eg, http://ihaveaclue.$ISP/enable?services=smtp or similar on reconnect doesn't have the necessary clue.) On a related tack, I am seriously considering writing to the appropriate government ministers and suggesting that, as part of their proposed anti-spam legislation, a legal duty be placed on people not to connect/allow to remain connected an insecure/0wned/infected system under their control. With the first breach resulting in mandatory disconnection of the system from the network, not to be reconnected until person had completed a course on "network security" and had their machine certified "cleaned up" by someone appropriate. Subsequent breaches resulting in that and fines and/or longer periods of mandatory disconnection. IMHO such insecure/0wned/infected systems are a nuisance (in the legal sense of the word) and thus the owners of them should be responsible for the damage they cause. Ewen

4 3

Google and MaxNet
by Cameron Kerr 09 Jun '04

09 Jun '04

Hello, is anyone getting `Your client has issued a malformed request' messages from Google when connected via MaxNet? I've tried with lynx, links, and Firefox. I'm getting these messages, not because my client is issuing a malformed request, but rather that MaxNet's transparent proxy is serving this. tcpflow reports the following in the response. Note the line that says `HIT' in it. HTTP/1.0 400 Bad Request Date: Tue, 08 Jun 2004 22:59:04 GMT Content-Type: text/html Server: GFE/1.3 Cneonction: Close <=== Hmmm, this is interesting. Content-Length: 1207 Age: 40 X-Cache: HIT from proxy2.akl1.maxnet.net.nz Connection: close This doesn't always happen under Firefox (it happened the first time, but after that, it worked ok. I'm guessing it's because of the short lifetime (Age: 40 seconds). Can anyone duplicate this? Just to be safe, I've disabled the transparent proxy here, and have no other proxy configured (direct connection). This was happening last night too. -- Cameron Kerr cameron.kerr(a)paradise.net.nz : http://nzgeeks.org/cameron/ Empowered by Perl!

12 22

Re: [nznog] Zombies
by neil gardner 09 Jun '04

09 Jun '04

To be fair though, in mentioning an opt-out system I was thinking of clueful users deliberately running their own servers that would notice within minutes that port 25 had been blocked - not the SMTP AUTH crowd which may be significantly less clued up and not notice why their previously working 'Email Thingy" isn't working anymore. I had the further thought that you could use some higher layer content inspection device to drop all SMTP traffic unless it's either to your own SMTP IP, or part of an SMTP AUTH conversation but that is way less than trivial as you'd need to completely proxy the sessions up until SMTP auth at least.... I'd say it was do-able, but not trivial. Cheers - N. >>> Nathan Ward <nznog(a)daork.net> 10/06/2004 2:50:13 p.m. >>> On 10/06/2004, at 2:43 PM, Perry Lorier wrote: > This breaks SMTP AUTH for people who send mail via another machine on > the Internet, because that host is the only one allowed to send with > SPF. > You conveniently ignored the opt-out system he mentioned. However for such an opt-out system to work, your packet filters would have to adjust themselves as dynamic users log in and out. That would suck. -- Nathan Ward _______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

4 3

RE: [nznog] Zombies
by Gordon Smith 09 Jun '04

09 Jun '04

The problem with that would be dynamically assigned addresses, unless you included a lookup being done against your radius databases by the SMTP proxy, so that the user can be identified. Possibly, a better solution would be to analyse netflow data as it hits the collectors, and identify offending machines that way. By using netflow, you can extend detection of worms, etc to include all ports, rather than just port 25 Doing dynamic throttling of infected connections is much more difficult - there's not a large range of equipment out there that can do that reliably, although the Juniper ERX does spring to mind :-) But that's not a box you'd find in most ISPs Cheers, Gordon > -----Original Message----- > From: neil gardner [mailto:neil.gardner(a)alliedtelesyn.co.nz] > Sent: Thursday, 10 June 2004 3:20 p.m. > To: nznog(a)list.waikato.ac.nz > Subject: Re: [nznog] Zombies > > > OK, serious thought here... Bear with me... I may be a) Describing an > existing system or b) way off base. > > Transparent SMTP proxy intercepts all messages and maintains a running > total of emails sent per source IP. This running total is actually > stored as a time distribution (ie. 10 emails in 10 minutes, > then none in > 10, then 100 in ten etc) > > Rulebase periodically examines this time distribution for each source > IP. Something like > > If last 3 matches are >10, >10, >10 then block for 30 minutes per > strikeout(sourceIP) and increment strikeout(sourceIP) by one. > > After expiry of downtime, open them up again. After (say) 90 > minutes of > non-abhorrent behaviour, decrement strikeout(sourceip) by one. > > If strikeout(sourceIP) gets to threshold (Say 10) then email user and > 48 hours later disable account unless they contact helpdesk > and claim to > have addressed the situation. > > You probably want cleverer rules to identify bursts of email, and > possibly a training period with rules referrent to earlier > data from the > same source > - this should be able to reliably identify legitimate high volume > users - especially if you build in a time of day factor. > > This should quickly throttle zombies and disable the account if not > fixed, but allow for brief infections to be throttled, and if > fixed, NOT > need manual intervention to unblock. It may impact some legit heavy > email users, but mail will bounce, and it should fix itself. > > Just a thought. > > Cheers - N. > >

3 2

RE: [nznog] Zombies
by Gordon Smith 09 Jun '04

09 Jun '04

Neil gets my vote. This is by far the best solution I've seen in years :-) And it's extendable to other protocols as well..... Cheers, Gordon

1 0

Re: [nznog] Zombies
by neil gardner 09 Jun '04

09 Jun '04

Point taken. OK, revised suggestion is this: Cut port 25 entirely, when people call to complain, press a button to play this recorded message. "Dear customer, in order to combat the Child pornography, Terrorist communications and Spam being sent by infected computers - WHICH YOU MAY HAVE! - we have blocked all popular email programs form communicating to the Intarweb thingy. In future, please feel free to use http://webmail.isp.co.nz for your email requirements. If on the other hand you support Child Pornography, are a member of an active Terrorist Cell, or wish to get a b1gger p3nis now!, then please press (1) and we'll unblock your evil Email connection." Cheers - Neil G ps. Sorry - I'll get back ontopic by about Tuesday week. :-) >>> Juha Saarinen <juha(a)saarinen.org> 10/06/2004 2:58:56 p.m. >>> neil gardner wrote: > To be fair though, in mentioning an opt-out system I was thinking of > clueful users deliberately running their own servers that would notice > within minutes that port 25 had been blocked - "Hey Mr/Ms Helldesker, how come nobody can send us mail anymore?" -- Juha NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

1 0

Re: [nznog] Zombies
by neil gardner 09 Jun '04

09 Jun '04

>>> Simon Lyall <simon(a)darkmere.gen.nz> 10/06/2004 1:46:38 p.m. >>> >It looks like there are lot of NZ based Zombies sending very large amounts >of Spam ( much of it in German) to some ISPs. Would it not be relatively easy to drop all port 25 traffic unless it was directed at your own SMTP server? * (Or are the Zombies sending through your SMTP anyway?) Cheers - N. * - Allow an opt out for anyone that even notices - if they notice, they won't be part of the problem. NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

3 2

Trade Me changes
by paul gold 09 Jun '04

09 Jun '04

Please be advised that the Trade Me group of sites are currently undergoing an addressing update. DNS records were changed last night, and all routing for our new address ranges continues to be via our upstream AT&T. Major sites affected; trademe.co.nz oldfriends.co.nz safetrader.co.nz findsomeone.co.nz (findsomeone.com.au, findsomeone.ca and personalstuff.co.nz plus other minority 'skins') circleoffriends.co.nz (circleoffriends.com.au) We are also physically moving the 'farm', which will introduce a number of interruptions and periods of downtime between Thursday 2300 and Friday 0700.

1 0

RE: [nznog] Google and MaxNet
by neil gardner 09 Jun '04

09 Jun '04

>>> "Craig Spiers" <craig(a)concept.net.nz> 10/06/2004 11:10:01 a.m. >>> >Heh.. Telecom ADSL's charges are for the data traveling across the telecom >data network, regardless where its come from.. This is a different situation >altogether.. You know what? I can't figure out if you're joking or serious :-) Really! N. NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

1 0

RE: [nznog] Google and MaxNet
by Regan Murphy 09 Jun '04

09 Jun '04

> In the past, caching has made some sense for accelerating > content to clients. I like it when I download a service pack from microsoft (for example) and the 100MB comes in at 20Mbps instead of the 512kbps that my international pipe is limited to. The fact that it may only technically be local traffic but I get charged for international traffic makes no difference because I would have no choice but to be paying international rates if there was no cache. When security/antivirus vendors release patches and updates is there a large increase in international bandwidth use or do the caches assist there? Caching must also be good for sites where there limited bandwith is at the server end too. -- Regan

4 5

RE: [nznog] Google and MaxNet
by Gordon Smith 08 Jun '04

08 Jun '04

Hahaha... but they were better than some of the other commercial offerings out there. I have heard good things about the more recent squid code (haven't tried it though) - but who wants to maintain bypass lists just because users want to use dodgy web software :-( > -----Original Message----- > From: Tony Wicks [mailto:nzog(a)road.gen.nz] > Sent: Wednesday, 9 June 2004 1:23 p.m. > To: 'David Fox'; 'Steve' > Cc: 'NZ Network Operators Group' > Subject: RE: [nznog] Google and MaxNet > > > Well, maybe it's just that Cacheflows suck ? > > -----Original Message----- > From: David Fox [mailto:foxy(a)morenet.net.nz] > Sent: Wednesday, 9 June 2004 1:12 p.m. > To: Steve > Cc: NZ Network Operators Group > Subject: RE: [nznog] Google and MaxNet > > > I agree wholeheartedly with Steve comments below, the cost saving is > negligible while the required resource to administer was > high. I redeployed > my cache to a new position in our network: it is working very > well as a > footrest under my desk. > > I found removing the cache actually improved overall performance, my > customers - and even managers - noticed the improvement ... > some of them > even rang to say Thanks (that makes a change!) > > David >

1 0

RE: [nznog] Google and MaxNet
by Gordon Smith 08 Jun '04

08 Jun '04

I don't think its so much particular brands that are an issue, but rather the sheer volume of poorly designed websites. With the popularity of MS Word & Frontpage, there seems to be a lot of ignorance relating to the W3C standards. In the past, caching has made some sense for accelerating content to clients. As far as bandwidth savings go, I've never seen that achieved to the point that it made them worthwhile in that role. They can improve performance as perceived by the customer, but bypass lists rapidly become unwieldy when trying to compensate for sites that don't adhere to standards. Overall, in terms of TCO, they're probably not worth having in place any more Cheers, Gordon > -----Original Message----- > From: Juha Saarinen [mailto:juha(a)saarinen.org] > Sent: Wednesday, 9 June 2004 1:13 p.m. > To: Alastair Johnson > Cc: NZ Network Operators Group > Subject: Re: [nznog] Google and MaxNet > > > So which caches have people had success/hassle with? > > -- > Juha

1 0

Re: [nznog] Google and MaxNet
by neil gardner 08 Jun '04

08 Jun '04

>>> Joe Abley <jabley(a)isc.org> 9/06/2004 12:52:46 p.m. >>> >Akamai has nodes in New Zealand, though. Or are you suggesting that NZ >ISPs routinely cache domestic content as well as content from overseas? >Joe Well they might have to start caching Domestic traffic shortly if Peering goes the way the big players would like it to go. Cheers - N. NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

1 0

Transparent Caching (was Google and Maxnet)
by Templeman Colin 08 Jun '04

08 Jun '04

A good topic! I 'discovered' transparent proxies recently when I was trying to resolve a problem I had obtaining security patches from security.debian.org via http. Other than having recently changed to cable broadband via Telstraclear, nothing (including apt sources) was altered, yet I was getting timeouts. After a couple of weeks of this, I eventually provided a comparison of latency output from 'ping' vs 'hping' to security.debian.org before the call was tacitly acknowledged and the problem silently fixed. I agree with Joe that they are an overhead when it comes to managing exceptions. -- Colin Templeman Security Analyst IT Risk and Security New Zealand Post Ltd This email with any attachments is confidential and may be subject to legal privilege. If it is not intended for you please reply immediately, destroy it and do not copy, disclose or use it in any way.

1 0

.nz Registry Directors
by Keith Davidson 08 Jun '04

08 Jun '04

Hello, Please pardon the intrusion to the NZNOG list. InternetNZ's subsidiary company, New Zealand Domain Name Registry Ltd, (NZRS) is responsible for operating the registry for the .nz namespace. This Company is governed by a board comprising 5 directors. Two directors retire by rotation at the Annual Meeting on 30 July 2004. Both retiring directors have declared themselves available for re-election. InternetNZ wishes to invite applications from others who believe they may have specific skills that would be advantageous to the Company. A full description of the position and skill sets being sought is available at www.internetnz.net.nz/public/registry/reg040606directors.html. Applications close on Monday 14 June 2004 (PLEASE NOTE - the web page states 11 June, but is in the process of being amended), and should be addressed to president(a)internetnz.net.nz or President, InternetNZ, P O Box 11-881, Wellington. Thanks Keith Davidson President - InternetNZ

1 0

Team Cymru Darknet Project
by Rob Thomas 03 Jun '04

03 Jun '04

-----BEGIN PGP SIGNED MESSAGE----- Hi, team. We are happy to at last release some documentation on Darknets and the Team Cymru Darknet Project. You will find the details at the following URL: <http://www.cymru.com/Darknet/> We hope this helps you to gain greater insight into your networks and the flows on them. Comments, feedback, and coffee are always welcome! :) Thanks! Rob, for Team Cymru. - -- Rob Thomas http://www.cymru.com ASSERT(coffee != empty); -----BEGIN PGP SIGNATURE----- Version: PGP 6.5.2 iQCVAwUBQL/XxlkX3QAo5sgJAQFBLwP/akP6qwmfg0y20wPRmX0QuOENh0NU4fhN N7LFdPqnYU3jJv6MgAR7J/u2snLmpC1RLZLkSvEFx4OtiYqqdVci4/9jT4D+M9nU qxERosZJhS65zZCExMEMJLEATWpr29pONEEBFbHBZgNYMDhee+fgDB6/GaEqY/Cr Y0qNvD0KRs4= =mm6q -----END PGP SIGNATURE-----

2 1

Queenstown
by K. Anand Lal 02 Jun '04

02 Jun '04

Hi, ihug is looking to upgrade it's Queenstown POP and as part of the upgrade we are considering a physical move to a better site. Is anyone aware of a telehousing facility located in central Queenstown? We will supply our own UPS and 19" enclosure. Appreciate any help. -- K. Anand Lal ph: +64.9.3592767 Engineering Manager fax: +64.9.3581518 The Internet Group (ihug) Ltd. noc: +64.9.3592708

3 2

The New Zealand Network Operators' Group
by Donald Neal 31 May '04

31 May '04

The New Zealand Network Operators' Group The New Zealand Network Operators' Group (NZNOG) has no king, president or formal membership. At present it consists of the subscribers to this mailing list, which anyone is free to join. Meetings See http://www.nznog.org/ for details of the January 2004 conference, which drew something over 120 participants. We are now seeking to confirm arrangements for a similar gathering to be held in the first week of February 2005. Also see http://auckland.thursdaynightcurry.com/ if you live in or near Auckland. Operators' Contact List See http://www.usenet.net.nz/noc/ for operational contact details for most New Zealand ISP's. These are intended for use by other network operators rather than by most customers. Internet Exchanges See http://www.ape.net.nz/ for details of the Auckland Peering Exchange and those connected there. See http://www.wix.net.nz/ for the Wellington Internet Exchange. About This Mailing List This list has around 650 addresses subscribed. You may only post to the list from a subscribed address. A number of people subscribe addresses which do not receive email purely to allow posting from them. The NZNOG mailing list is provided through a server at The University of Waikato, and is administered by employees of Alcatel NZ Ltd, Citylink and ihug Ltd. None of these organisations, nor any administrator, is responsible for its content. NZNOG Mailing List Acceptable Use Policy The NZNOG mailing list exists to provide a forum for the exchange of technical information and the discussion of implementation issues that require cooperation among New Zealand network service providers. In order to continue to provide a useful forum for discussion of relevant technical issues, users of the list are asked to respect the following guidelines. 1. Discussion will focus on Internet operational and technical issues. 2. Discussion related to meetings of network service providers is appropriate. 3. Discussion unrelated to these topics is not appropriate. 4. Postings to multiple mailing lists are discouraged. 5. Postings that include foul language, character assassination, and lack of respect for other participants are unacceptable. 6. Blatant product or service marketing is unacceptable. 7. Postings of a political, philosophical or legal nature are discouraged. 8. Postings to the list should be in ASCII or MIME encoded as text/plain. Attachments should not be sent to the list. To present a document, a suitable URL may be referred to. For documents of general interest, the use of proprietary file formats is discouraged. 9. Breaches of list etiquette should be dealt with privately with the offending list user, and should not result in complaints being sent to the list. 10. A person repeatedly breaching list etiquette shall receive warnings from the list administrator. A further breach after the second such warning within thirty days shall result in the offender being unsubscribed from the list. Other action may also be taken to block postings to the list by the offender. Any such unsubscription is to be immediately announced to the list. Mailing List Archives A full archive is available at http://list.waikato.ac.nz/archives/nznog/ Any message sent to the list will be archived and made available on the web automatically. Changes are not made to the archive on request, though the administrators remain happy to assist the Office of the Privacy Commissioner should any complaint be laid with that office. One way to search the archive is to use google and prefix your search with site:list.waikato.ac.nz . Subscribing to the List See http://list.waikato.ac.nz/mailman/listinfo/nznog to subscribe. Donald Neal NZNOG Mailing List Owner/Administrator/Muggins ------------------------------------------------------------------------------ "This communication, including any attachments, is confidential. If you are not the intended recipient, you should not read it - please contact me immediately, destroy it, and do not copy or use any part of this communication or disclose anything about it. Thank you. Please note that this communication does not designate an information system for the purposes of the Electronic Transactions Act 2002." ------------------------------------------------------------------------------

1 0

Re: [nznog] TC & T Possible Peering changes
by neil gardner 30 May '04

30 May '04

>>> Simon Lyall <simon(a)darkmere.gen.nz> 28/05/2004 2:24:47 p.m. >>> No, they'll change their terms and conditions so that National traffic is no longer free and is instead part of your quota. [snip] Yeah, I knew this would probably be the way they handled it. Thanks Telecom and TelstraClear, for between you managing to keep cheap high speed Internet a far off dream. Cheers - N. NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

3 2

RE: [nznog] TC & T Possible Peering changes
by Gordon Smith 30 May '04

30 May '04

One thing that noone seems to have noticed yet, is that using TCNZ for transit in this way is virtually impossible due to the manner in which they route traffic. TCNZ use BGP communities to tag and filter, as did Clear before them. So ingress and egress paths are filtered at the exchange points. I don't know if Telecom have started using communities to manipulate traffic paths yet, but if they haven't, I'd say it's only a matter of time. The peering argument is a political one rather than a technical one. In my opinion, "unauthorised transit" is nothing more than a convenient excuse. If it did actually happen, perhaps they'd be better odd re-visiting filtering policies. Just my .02c worth :-) Cheers, Gordon

2 1

Re: [nznog] Indian government plans Internet exchanges
by Keith Davidson 28 May '04

28 May '04

Others wrote: > > Someone with some spare hands on their time needs to lobby > > the government to make peering in NZ legislation. > > > > Although TNZ and TCNZ will be wearing a lot of the network > > costs in terms of trunking traffic up and down the country, > > they also have most of the customers - so I think that's fair > > and justified. > > Yes it is good idea. > > So why not also make the Govt responsible for the trunking between > peering sites as well? > ISPs could still organise private peering for priority traffic in this > scenario. Is this an issue that nznog feels InternetNZ should pick up on? Keith Davidson President - InternetNZ

4 7

RE: [nznog] TC & T Possible Peering changes
by Donald Neal 28 May '04

28 May '04

> On Behalf Of Russell Fulton > Sent: Friday, 28 May 2004 1:56 p.m. > To: nznog > Subject: Re: [nznog] TC & T Possible Peering changes > > > On Fri, 2004-05-28 at 13:46, Don Gould wrote: > [...] > > * Why don't the IX have clear policies on this issue? > > Perhaps a better question is "When are the IXs going to > establish clear > policies on this issue?" > > Russell. Perhaps they already have a clear (no pun intended) policy - they're not responsible for what routes you advertise. - Donald Neal All opinions mine only. Donald Neal | Alphabet - ancient Technical Specialist | Greek lottery. Operations Engineering | [Martin Neal] Integration & Services Division +----------------------- Alcatel NZ Ltd - Telecom's network operations manager ------------------------------------------------------------------------------ "This communication, including any attachments, is confidential. If you are not the intended recipient, you should not read it - please contact me immediately, destroy it, and do not copy or use any part of this communication or disclose anything about it. Thank you. Please note that this communication does not designate an information system for the purposes of the Electronic Transactions Act 2002." ------------------------------------------------------------------------------

3 2

RE: [nznog] Indian government plans Internet exchanges
by Seeby Woodhouse 28 May '04

28 May '04

What a good idea. Someone with some spare hands on their time needs to lobby the government to make peering in NZ legislation. Although TNZ and TCNZ will be wearing a lot of the network costs in terms of trunking traffic up and down the country, they also have most of the customers - so I think that's fair and justified. If they are worried about people using their networks as a free national backbone via the peering points, there are ways to allow peering but prevent being used as free transit... Cheers Seeby -----Original Message----- From: Andy Linton [mailto:asjl(a)citylink.co.nz] Sent: Friday, May 28, 2004 9:33 AM To: nznog Subject: [nznog] Indian government plans Internet exchanges This URL points to the incoming Indian governments 10 point agenda form IT - http://www.ciol.com/content/news/2004/104052609.asp In particular, see point 5: 5. Plan to connect all ISPs in India to a national internet exchange in order to achieve efficient internet traffic routing, cost reduction and improve quality of service for the Internet users in India. Four Internet exchanges points to be fully operationalised. Small and big ISPs to route their traffic through this exchange to ensure the security of domestic traffic. Currently, 25 ISPs are connected. ---- Note the reasons for doing this especially "improve quality of service for the Internet users". _______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog

7 6

RE: [nznog] TC & T Possible Peering changes
by Barry Murphy 27 May '04

27 May '04

Ok, But I still fail to see how this works. 1./ I buy my international pipe through telstra 2./ I buy my national pipe through telstra 3./ I try connect to nzdating.com but because iconz doesn't per with telstra, I use international and not 'national' and my pictures load SLOWLY!!! *phones helpdesk to complain about slow browsing* I'd freak. Whats T&T going to do to provide 'national' connectivity? Is it only national if the remote ISP peers with T. Is national defined as being within T's network only? That sucks BIG time. Somebodys been playing with their calculator to long and doesn't understand the internet to be able to make decisions like this. B My thoughts alone. -----Original Message----- From: Craig Spiers [mailto:craig(a)concept.net.nz] Sent: Friday, 28 May 2004 4:02 p.m. To: 'Barry Murphy'; 'Don Gould'; 'neil gardner' Cc: 'nznog' Subject: RE: [nznog] TC & T Possible Peering changes I think this is a stupid idea, how would you control who pays for what percentage of the link? What about the larger networks who utilise 50% of the shared link? Better to just get your own link and be done with it.. :) -----Original Message----- From: Barry Murphy [mailto:barry(a)unix.co.nz] Sent: Friday, 28 May 2004 3:59 p.m. To: 'Don Gould'; 'neil gardner' Cc: 'nznog' Subject: RE: [nznog] TC & T Possible Peering changes Why would everyone have to peer with T&T? 1./ Why don't all the ISP's currently peering at APE & WIX chip in a little and purchase a link between APE and both T or T. 2./ Assuming the traffic is flat rate and they don't charge you big time for the link, ISP's have domestic BW again 3./ Any traffic coming from T or T should go international, until such time as T or T purchase links to APE or your provider. I know this scenario is going to use up international bandwidth, but hey, outgoing bandwidth isn't as important as incoming. T&T customers will suffer the most. Barry -----Original Message----- From: Don Gould [mailto:don(a)bowenvale.co.nz] Sent: Friday, 28 May 2004 1:46 p.m. To: neil gardner Cc: nznog Subject: Re: [nznog] TC & T Possible Peering changes That's a really good question... 'bring it on'... Small claims wouldn't take much to defend and I'd suggest that's a fight the isp wants to loose (without it costing them to much in traffic so a smaller customer would be prefered). Start getting cases like this turning up in the court system and you've got leverage for a number of government agencies to take actions. However what's this question really about? Like you Neil I've been reading thru the endless banter and a few other questions come to mind... * Who are these ISPs that are using T&T for transit that shouldn't be? Why are we not also focusing on the limited number of these folk who are wrecking things for everyone else? ...or am I about to be told to shut up and keep quiet because I'm showing ignorance by not knowing that everyones doing it?! * Why don't the IX have clear policies on this issue? Why is the core of each IX not dealing to this? IX are generally a community venture, why haven't the IX set up to protect the community interest against these rogue users? Don't get me wrong, I'm not picking sides here, I've just been following the discussion and it seems to me that people are keeping quiet on some issue - perhaps I'm just stating the obvious and as a list newbi I'm just not up with all the politics yet? Someone called for someone with time to take the issue up with some lobying.... I'd put up a hand and have a look but as I see it at present the arguments suggest that the industry got what was coming to is... please don't abuse and flame me for that comment, I'm just saying it as I see it based on what's been presented in this discussion todate. Cheers Don _______________________________________________ NZNOG mailing list NZNOG(a)list.waikato.ac.nz http://list.waikato.ac.nz/mailman/listinfo/nznog

3 2

TC & T Possible Peering changes
by neil gardner 27 May '04

27 May '04

Goodness there has been a lot of stuff to wade through. I'm quite familiar with the benefits of peering and the (revenue) reasons that the big players want to change, but could someone please definitively answer this question... Will the proposed changes by Telstra mean that a large number of people currently paying different rates for National and International traffic will suddenly find that their ISP starts defining an awful lot more geographically national servers as International traffic? If this is indeed the case, have the various ISPs gotten their T&C well enough defined to defend against someone going to the small claims court and disputing a bill where traffic is obviously from a national server, even if it has transited an international link - since the ISP (may have) made a deliberate choice to route that traffic in a non-optimal fashion? Cheers - Neil G NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

8 8

Re: [nznog] TC & T Possible Peering changes
by neil gardner 27 May '04

27 May '04

>>> Juha Saarinen <juha(a)saarinen.org> 28/05/2004 2:06:47 p.m. >>> You're kidding right? We have T&Cs that speak of "best effort service" with no guarantee as to any aspect of delivery. I kid not... This quote is taken (almost at Random) from Paradise... "JetStart Traffic is calculated depending on the location of the site the traffic is coming from or going to. Any traffic to or from New Zealand sites (national traffic) is charged at 1/10th of the normal traffic charge, while traffic to or from a site outside New Zealand (international traffic) is charged at the full traffic rate." That tells me (in my mind) that traffic from a site in New Zealand will be charged at 1/10th of the normal rate. No normal home user should be expected to understand that just because of a spat between 2 National ISPs, Traffic from me here in Chch to a server in Wellington might go via Sydney and therefore be classed as International Traffic. Interestingly, as Paradise have that written right now, it doesn't matter. The way they have used the word "Site" in that context would clearly convey to a reasonable person that they mean the end server, not the last or most remote hop... So even if my traffic to and from Trademe goes via Timbuktu, the "Site" is Trademe, and it's a "New Zealand sites (national traffic)" (sic) Cheers - N. NOTICE: This message contains privileged and confidential information intended only for the use of the addressee named above. If you are not the intended recipient of this message you are hereby notified that you must not disseminate, copy or take any action in reliance on it. If you have received this message in error please notify Allied Telesyn Research Ltd immediately. Any views expressed in this message are those of the individual sender, except where the sender has the authority to issue and specifically states them to be the views of Allied Telesyn Research.

3 2

RE: [nznog] Indian government plans Internet exchanges
by Regan Murphy 27 May '04

27 May '04

> Someone with some spare hands on their time needs to lobby > the government to make peering in NZ legislation. > > Although TNZ and TCNZ will be wearing a lot of the network > costs in terms of trunking traffic up and down the country, > they also have most of the customers - so I think that's fair > and justified. Yes it is good idea. So why not also make the Govt responsible for the trunking between peering sites as well? ISPs could still organise private peering for priority traffic in this scenario. -- Regan

3 2

RE: [nznog] Indian government plans Internet exchanges
by Gordon Smith 27 May '04

27 May '04

> -----Original Message----- > From: Craig Whitmore [mailto:lennon(a)orcon.net.nz] > Sent: Friday, 28 May 2004 12:33 p.m. > To: 'NZNOG Mailing List' > Subject: RE: [nznog] Indian government plans Internet exchanges > > > Why don't they just (like a lot of other people do), Advertise their > Auckland networks via APE only and their Wellington Networks > via WIX only. > Then they wouldn't be providing transit down the country. > > Advertising the Entire Clear Network via APE and the Entire > Clear Networks > via WIX is the problem. > Thanks > Craig > The use of BGP communities would help solve their problems regarding routing ingress and egress points. Pity it isn't more widely used..... Transits can be filtered by community tag within an AS, and most major vendors support using bgp communities in this way

1 0

Indian government plans Internet exchanges
by Andy Linton 27 May '04

27 May '04

This URL points to the incoming Indian governments 10 point agenda form IT - http://www.ciol.com/content/news/2004/104052609.asp In particular, see point 5: 5. Plan to connect all ISPs in India to a national internet exchange in order to achieve efficient internet traffic routing, cost reduction and improve quality of service for the Internet users in India. Four Internet exchanges points to be fully operationalised. Small and big ISPs to route their traffic through this exchange to ensure the security of domestic traffic. Currently, 25 ISPs are connected. ---- Note the reasons for doing this especially "improve quality of service for the Internet users".

5 4

netlantis news
by Pascal Gloor 26 May '04

26 May '04

Hello ppl, as you probably have noticed, netlantis is down since a while. Netlantis had critical performance problems and we decided to re-write some of the core scripts to improve the DB update. While doing this we had some new ideas about how to process the updates and much more has been re-written. There are no new features for the users, just the core-system works on a different way and all the BGP updates queuing problems will be gone. "What's new then ?" netlantis is running now since several weeks its own BGP daemon called (guess) nbgpd. So far it works well and has pretty nice performance (able to handle up to 500 full BGP sessions (tested up to 100)). "nbgpd" is dumping updates into files (per peer, per timestamp) in its own (simple) binary format. The updates are then sent to "ndb" (netlantis DB manager) which is a fast "routing table status file updater". This also runs since several weeks. the last difficulty is to convert this "ndb" format for the SQL database (to handle all the web/whois/telnet queries). So far we wrote the MySQL (MyISAM format) data file properly. "So? why isnt it up yet ????" We still have a last part to finish, writing the MySQL (MyISAM format) index file. Once this step completed, netlantis will come back up! "What's the difference for the lambda user??" Reliable informations! Until now, "global cisco bug upgrade worldwide in the world" made netlantis to go booom due to the high load of bgp updates. This will no more happen. We expect to have a delay to "real-time" of about 10 to 15 minutes, with few or many BGP updates, it wont matter! "Well, I dont care about your internal stuff, when will it be back????" We expect to bring netlantis back during the next week. We will inform you then. Best Regards, Pascal Gloor _ ______________ ___ _ ________________ / |/ / __/_ __/ / / _ | / |/ /_ __/ _/ __/ / / _/ / / / /__/ __ |/ / / / _/ /_\ \ /_/|_/___/ /_/ /____/_/ |_/_/|_/ /_/ /___/___/.org ... where all the routes meet

1 0

Location of Tangent/Vector Fibre in Auckland, Hopetoun Street
by Paul Adshead 26 May '04

26 May '04

Does anyone know the current extents of the Tangent/Vector fibre in Auckland? More specifically how far away is it from the Ponsonby end of Hopetoun Street. I'm trying to find out how far away it is and whether it's an option or not, but Vector are being difficult... Regards, Paul Adshead Network & Systems OTL Software Ltd www.otl.co.nz

4 4