- NANOG - lists.nanog.org

RE: ISP Policies
by Michel Py 09 Sep '04

09 Sep '04

> Tulip Rasputin wrote: > That's why i explicitly asked for some "social/political/etc." > reasons where an ISP may not want his traffic to traverse some > particular AS number(s). Something which is beyond BGP to > determine as of now ! :-) FWIW, this is exactly how I understood the question. It's all about "non-BGP" issues. > I believe with the responses that i received both on the list > and offline, that it is indeed quite normal for ISPs to filter > routes based on the AS Paths for 'other' reasons. Reasons, > quite beyond BGP as a protocol to handle! And this can happen, > when an ISP doesnt want its traffic to traverse some AS(es). I'm not sure I agree with "normal", but it is common practice indeed, a significant part being in a <cough> grey <cough> area, and notice that nobody dared to post the reasons on the ML. Trying to stay intellectually honest, there are "good" and "bad" reasons for it. Using my well-known politically incorrect bluntness, I would say this (words borrowed from several people) (disclaimer #2: this is somehow exaggerated, but here it is anyway): God has given men a brain and a penis, but not enough blood to operate both of them at the same time. Although there are exceptions, when the blood flows to the brain, men use BGP; when the blood flows to the penis, men manipulate the AS_PATH and/or create route-maps :-D Michel.

1 0

ISP Policies
by Tulip Rasputin 09 Sep '04

09 Sep '04

Hi, I have a general policy question. Do the ISPs ever look for some particular AS number in the BGP AS_PATH and then decide what action/preference/priority they need to take/give based on the AS number(s) present in the BGP AS_PATH_SEQ/SET? For instance, does it happen that an ISP receives some BGP paths, but because of some political, social, economical, DOS attack, etc. reasons decides that it doesn't want to accept this path because some particular AS number is present in the BGP UPDATE. Basically, it doesn't want *its* traffic to flow via that particular AS number(s). Or, if there is a mutual disagreement between two ISPs, and one doesn't want his traffic to traverse the other's AS number. Does this sort of thing ever happen? Are such restrictive policies normal in the ISP/IX scenarios? Thanks, Tulip

5 5

AT&T Wireless IP Clue
by Steve Rubin 08 Sep '04

08 Sep '04

Does anyone have an IP clue contact at AT&T Wireless? They have some sort of filter blocking most (all?) of 69/8 to their SMS email gateway, and even the AT&T Wireless NOC can't seem to find the problem (800-832-6662, option 6, ticket #1087071). I was leaning towards 69/8 bogon filters until I noticed that twice a day for about 30 minutes everything works.

1 0

Re: Spammers Skirt IP Authentication Attempts
by Robert Bonomi 08 Sep '04

08 Sep '04

> From owner-nanog(a)merit.edu Wed Sep 8 12:05:02 2004 > To: nanog(a)merit.edu > Subject: Re: Spammers Skirt IP Authentication Attempts > From: Paul Vixie <vixie(a)vix.com> > Date: 08 Sep 2004 16:59:51 +0000 > > > vgill(a)vijaygill.com (vijay gill) writes: > > > ... That means that if I do get a mail purporting to be from citi from > > randomgibberish, I can junk it without hesitation. > > agreed, that is what it means. > > however, and this is the important part so everybody please pay attention, > if you can junk something "without hesitation," then spammers will stop > sending that kind of "something." they make their money on clickthroughs, > final sales, and referrals, which translates to one thing and one thing > only: "volume." if the way to keep their volume up means "put SPF metadata > in for the domains they use" or even just "stop forging mail from domains > that have SPF metadata" then that is exactly what they will do. guaranteed. > > there's a bet here. you could bet that by closing off this avenue, SPF will > force spammers to use other methods that are more easily detected/filtered, > and that if you play this cat&mouse game long enough, it will drive the cost > of spam so high (or drive the volume benefit so low) that it'll just die out. > I, for one, don't think that SPF is a FUSSP (tm vernon), or anything close to it. I _do_ think that it is _a_step_ 'in the right direction'. I'd *love* to see SPF-type data returned on rDNS queries -- that would practically put the zombie spam-sending machines out of business. SPF _can_ serve a 'useful function' in spam-fighting. As follows: SPF verification query gets returns one of three kinds of result: 1) MISMATCH on point-of-origin vs domain 'authorized' senders. *VERY* probably spam. Need a white-list check of the specific sender e-mail, and if that fails, an SMTP-session rejection is indicated. 2) MATCH on point-of-origin for domain vs domain 'authorized' senders. 'reliable' data-point that the domain owner 'authorized' the use of the domain-name. Now it makes sense to query an _internally_ _maintained_ database of 'familiar to me' domains, to see what types of prior mail from this domain have been seen. This is a much simpler, quicker, less CPU-intensive test than the 'full' set of spam checks. Match on 'known spammer' domain causes immediate SMTP-time rejection; match on 'known NON-ISP, non-spammer' domain and one is quite 'safe' in accepting the message without further checks. Messages from 'unfamiliar' domains, and/or 'ISP' domains, get the full spam-check treatment. 3) NO DATA. In this scenario, doing the full set of spam checks, is unavoidable. Unless you reject traffic based on the lack of SPF data; which is a non-starter strategy, until such time as SPF is near-universally deployed. If _nobody_ published SPF data, then the situation degenerates into case 3), as a worst-case scenario. This is no worse than the situation _without_ SPF checking. (For the nit-pickers, yes, it is a _little_ worse, by the overhead of the 100% no-constructive-date SPF queries.) If a 'non trivial' share of the incoming message traffic falls into case 1) and/or case 2), then the use of SPF is a net 'win' for the recipient. *IF* the time comes when SPF deployment is near-univeral, then case 3) drops out of the picture. _IN_THAT_CASE_, spammers pretty much _have_ to publish SPF data for their outgoing mail sources, to have any 'hope' of delivery. Whereupon, either they're sending from 'known spammer' domains, or the 'unfamiliar domain' handling kicks in. It aint the (or even 'a') FUSSP, But it is a _big_ win for places that handle large volumes of e-mail. For those big shops, it doesn't take long for a spammer domain to get out of the 'unrecognied' category, and into the 'known spammer' class. Whereupon, one SPF check, plus one internal database dip, and they can dump the mail as from 'known spammers'. The savings in system resources, by using such an approach on several _billion_ pieces of mail per day is definitely non-trivial. It takes a while for wide-spread acceptance/implementation, but when that state of affairs _is_ achieved, large-scale spammers have a serious problem: Messages claiming to be from sources lacking SPF validation get rejected. Messages with SPF-mismatch get bit-bucketed. Messages with SPF-validation from identified spammer domains get bit-bucketed What's an _honest_ spammer to do? <muffled snicker>

1 0

Re: Spammers Skirt IP Authentication Attempts
by Paul Vixie 08 Sep '04

08 Sep '04

> True, but bounces, and anything else with NULL return path, can be taken > care of with SRS. SRS is probably a higher pairwise deployment barrier than SPF. but in any case you should take this argument to the IETF MARID WG, since getting agreement on nanog@ (assuming it's possible) won't stop the SPF steamroller. > See: > > http://www.libsrs2.org/ > http://www.libsrs2.org/srs/srs.pdf > http://asarian-host.net/srs/sendmailsrs.htm > > And be happy, and realise "SPF is worthless" ;) SRS looks like a better technical solution than SPF, but it's less deployable. for one thing, There Can Be Only One SRS-like thing. there are already many SPF-like things, each with its own adherent-base, and there will be many more. > Is it really worth it for every domain owner on the planet (including > spammers!) to implement SPF records in DNS, and the resulting forwarding > breakage, simply to provide some fairly intangible "dilution protection" > for, primarily, the very small subset of widely-known domains out there? no. it's the same kind of cost/benefit assymetry as spam, where everybody has to pay a higher cost but only a few get a significant benefit from it. however, beta was better than vhs, too. and tully's is way Way better than starbucks. being better isn't as relevant as having better marketing. with microsoft backing SPF++ (is it "sender-id" now?), SPF will be widely deployed and the costs and benefits be damned. > > ... i'm glad that companies bigger and richer than i am find it in > > their own selfish best interests to push something like SPF -- that > > means it'll happen. ... > > Well that depends. At the moment it looks like the clients will > implement a standard that most of the servers will not! i've begun to hear privacy related concerns, as well. even with jim miller's MAIL-FROM proposal, there's a way to look at the DNS query stream and find out what servers are presently being spammed using your domain name as the source. this is an information leak but i'm willing to live with it. many MTA operators will not be willing to live with this. (maybe some large ones.) > > it's useful, just not for the advertised reasons, or a universal reason. > > Ah, absolutely yes. so, i'll take your "SPF is worthless!" statement under advisement.

1 0

summary (Re: OT- need a new GSM provider )
by Paul Vixie 08 Sep '04

08 Sep '04

i'd asked: > > Anybody had notable (good or bad) billing and/or customer service > > experiences with Voicestream or any other GSM provider with native > > coverage in the San Francisco Bay Area? many people said: > I think Voicestream and T-Mobile are the same company now. If you've > had problems with one, you'll likely have 'em again. now, most of the problems i'm having are due to statistical anomolies (in which BigTelCo is optimized for the average case, and i'm some kind of an exception). in that light, my problems are completely portable. however, AT&T is more statistically driven (that is, they've optimized their value proposition) than even t-mobile was. i may end up going back to t-mobile, since t-mobile appears to have lost the "race to the bottom" at least for now. you never know how good you had it until after you switch to AT&T, or something. i won't be switching to CDMA or any other non-GSM. just because us-gov deliberately allowed contractors to put closed, non-GSM systems into iraq and afghanistan does not mean i'm ready to be locked into old (bad) ways of doing my telco. until we have ubiquitous 802.11 so that i can roam with a SIP phone, GSM is the closest thing to "right". this isn't a technology problem -- GSM might be horrible under the covers, i don't know or care. if another multi-vendor international standard is proposed, i'll look at it. other observations (anonymized) that came in on this thread: --- > Bad news, in the bay area, T-Mobile == Voicestream. > You could go for Cingular, but you won't be any happier. The bad news > is that every GSM provider here just plain sucks. i'll stop including similar statements after this one. i'm struck by the fact that no matter what BigTelCo you ask about, you can hear a plethora of horror stories. somewhere, there may even be a happy AT&T customer or two. for the record, the best customer service experience i ever had was with NexTel, but since my office building kept moving in and out of the coverage region all day long, i had to switch to a company who could keep up with us. (i guess it's a san andreas fault thing.) so, the best provider might be the one with the best network, or the best prices, or the best customer service -- it's subjective. it does make me wonder if there's some kind of antitrust issue involved in the whole "race to the bottom" thing. the carriers are competing to see who can best-optimize their channels (spend the least, make the most) and the only reason they don't go out of business as a result is that every other provider is doing the same thing. if any provider were to "break ranks" and decide not to squeeze the margins quite so hard, they could make up by 10000X in volume what they "lost" by not constantly searching for their customers' collective pain threshold. but i'm bitter, and i digress. --- > I support about 15 cell users on a half dozen various providers, and > in dealings with support departments i've found T-Mobile to be the > least-horrible. No idea on your particular billing issues though, all > the folks there I have spoken with have had at least slight clue. > > I concur on avoiding AT&T like a plague. t-mobile's inability to let my phone ring through was the big issue. my call queuing system relies on timeouts to go to the next endstation. but t-mobile insisted on terminating every call that came to them, whether i was in coverage or not, whether i answered the call or not. i could choose between a voicemail answerbot or an error message, but they were unwilling to just let it ring. i got the idea that this was so they could collect call termination fees no matter what -- another "race to the bottom" idea. but it meant that my follow-me technology had to terminate in my t-mobile number, and t-mobile had to be configured to forward unanswered calls to my second, secret, hidden phone number, that continued my ring-queue and put my voicemail into my centralized inbox. and of course t-mobile charged me by the minute for the forwarded calls, even though they were local, and even though i was on the highest-priced possible calling plan. again, i'm a statistical anomoly for these people. they make their money from volume, and if i can't fit into their profile, they can't afford to try to please me. --- > Paul, you're pretty much left with cingular if you can't deal with > tmobile/voicestream and ATT. ATT will be merged in with cingular in a > few months. Other option is to get a Verizon samsung worldphone, which > will roam onto vodafone GSM internationally and us CDMA. That is sim > locked though. yeah, well, i'm done with sim-locked anything. > Also, do NOT get a triband phone, most triband phones are 900/1800/1900. > The largest amount of US GSM (both cingular and ATT) are implemented on > GSM 850. So you either want a quad band like the v600 or treo 600 which > can operate on the US (850/1900) gsm bands as well as the international > 900/1800 bands. that's important information, and sadly, it narrows the field by a lot. i really liked my treo-600 other than that it was so hideously large. a pity about it getting run over by a tractor like that. but it had quad-band. i will make sure that all four (850/900/1800/1900) bands are present in my next handset. --- > I've dealt with AT&T, T-Mobile, Verizon and Sprint around the bay area. > I'm never, ever doing any business with AT&T or T-Mobile again. I'm > happy with Verizon, and have been for years. I've been using Sprint > for about six months (a treo, for ssh on the road) and they've been > very easy and professional to deal with, no problems at all. i'm beginning to suspect that closed systems like sprint are going to give better domestic-US service. but that's like saying apple mac/os/x is a better environment for running msword. ("just because it's not windows doesn't mean they're not going to try really hard to lock you in.") --- > Verizon now has a worldphone that will roam onto vodafone GSM > internationally. Their rates don't appear to be too prohibitive. Though > if you are going to be calling a lot while abroad, I suggest picking up > an unlocked nokia 6310i and prepaid sims as you fly into airports. i'm beginning to wonder if buying a prepaid SIM once a month wouldn't give me the best price/performance even when i'm not travelling. my follow-me can cope with cellphone# changes once a month. the idea of not having a contract with any BigTelCo really appeals to me. even if it costs more. if their customer service and billing are terrible, which they will be, i can buy my prepaid SIM from a different company the next month. interesting. --- > If there ever actually is a GOOD cellphone provider, please let me know. > I've never found one that knew how to run their billing, or run a useful > customer service department. i think that an event of that magnitude (a company large enough to own global or even national infrastructure, but still competent at billing and customer service) would be so significant that you'd hear about it from others than just me. it's make the front page of every newspaper in the world. etc. --- > He's what I understand about T-Mobile/Voicestream: > > - T-Mobile used to be Voice Stream > > - T-Mobile in the Bay Area (and all of CA and NV as far as I know) is > resold Cingular > (http://www.gsmworld.com/cgi/ni_map.pl5?cc=us&net=we) > - This is changing as Cingular buys AT&T and sells off the redundant sites > to T-Mobile. > > From what I've heard and seen (going out to a few cell sites) AT&T > made some really short sighted mistakes in deploying it's network. > Like no migration path planed for GSM850. They even went as far (as > did T-Mobile) as selling handsets that either had only GSM1800/1900 > support or if they were tri-band were only GSM800, only useful in non > US markets (like Germany). Cingular on the other hand has been > selling GSM850 enabled handsets for years. In a good portion of the > Bay Area I getting both 850 and 1900 with the exception to parts of > the Sbay. The advantage of 850 over 1900 is of course greater > penetration into buildings and through trees. > > I have an unlimited data plan on Cingular for an extra $15 or so a > month. I get about 15-60kbps and 150-1500ms latency depending how > fast I'm traveling and signal strength. I use and love my Ericsson > T637. > > I'm not proud of it but until a better GSM provider is offered in CA I > have to admit I've been really happy with Cingular. I really have > tried in the past to hate them but after seeing other people have so > much trouble with T-Mobile and AT&T I'm forced to admit that they're > probably the best choice for me in this area. that's all very useful and would lead one to consider cingular. (i know this author, he's even more phone-destructive than i am.) --- > If your swapping phones definatly go with a Samsung E715 unlocked, > very very good phone in my opinion and I have never lost coverage in > the US or Europe (about to try it out in australia) since it only does 900/1900, and vijay and others have said that 850/1800 are also important, i think i'll forego this one. the NEC-515 is looking to be about right. ustronics sells them (unlocked) for $210. i've gotta have a flip phone, and i've got to be able to get it out of my pocket without hitting the dial-by-voice button (like on the moto-V600, ick.) the NEC-515 has java, so it's going to be slow, but at least it has no camera and no dial-by-voice. --- > > ... about to enter into another year long "you violated the > > agreement first" small claims battle, > > Been there. The court saga for only few $$$ is usually not worth your > time and any collection efforts for contested cellphone (and most > other telco) charges and contracts can be stopped with couple properly > written letters (its their job to go after you in court, not yours). i'm worried that they'll send me to collections and that i'll have to explain to the credit reporting companies that this overdue nonpayment was due to a dispute rather than flakiness. i'm still working that out with pac*bell (SBC) on an old MFN T-1 bill. this time i'm sending AT&T notice of their default so i can fax it to the credit reporting companies later on. not that any of them will care -- at this scale i'm just a statistic (and an anomolous one at that.) --- > http://www.gsmworld.com/roaming/gsminfo/cou_us.shtml sigh. none of the people who are small enough to send me an accurate invoice have any roaming partners in the san francisco bay area. i guess that's how the world operates. > But as far as I know T-Mobile and ATT are the only two nationwide GSM > providirs (Cingular too, but I hear it will soon be same as ATT). the thing about cingular is that it is actually not AT&T from an SG&A standpoint. it's possible that they'd be willing to pay an extra $0.05 per customer per month to keep their billing and customer service departments well staffed and well fed. (something AT&T would not and could not do.) i may give cingular a try. > In my opinion GSM is really overrated and not seriously well deployed > in US, consider CDMA providers, at least internet access would be > faster (and typically cheaper) if you're using smartphone. <rantish flame about US telecom backwardness elided> --- > They don't seem to care if you use your own phone, and they seem to lag > behind in some of the non-voice service offerings, but the covereage is > good, and the plans (as long as you get one of the ones over $50) are > pretty good minutes-wise. > > Their GPRS rates are kinda high but I think they have an all you can eat > now. i was never able to get gprs working on my win/2k laptop. maybe i'll try again -- it was one of the only reasons i had win/2k on that laptop. --- > Cingular has been pretty good to me customer service wise, though > their coverage is not quite as great as that of Verizon. It is > getting better since they are mergin with AT&T and are able to add > transmitters to AT&T's towers. Of course if you're sad about AT&T I > can't tell you what might happen when the merger is completed. i don't think it's a merger. others here have said that cingular is buying some of at&t's assets, so that at&t can go into the CDMA business. <http://www.attwireless.com/press/releases/2004_releases/051904.jhtml> says: Under the terms of the agreement announced on February 17, shareholders of AT&T Wireless will receive $15 in cash per common share, or a total of approximately $41 billion. so while it's a merger on paper, it's an acquisition on practice. i know which company's executive team will survive. cingular wanted the customer base and real estate assets, and merging was the way to buy those things without paying taxes on them. this is how acquisitions are done. no worries; the AT&T Wireless executive team can retire after this, and won't be running any important part of cingular (or anything else.) > BTW, did you know that comcast is blocking your mail: > > 553_Service_unavailable;_Client_host_[204.127.202.???]_blocked_using_reject-a > ll.vix.com;_reason_/_created actually, that's me rejecting mail that came from comcast, since they don't have the means to prevent their virus-ridden customers from operating spam relays. see www.vix.com/personalcolo/ for a much longer rant about this. --- > AT&T used to rip me a new one for intl->intl calls, but t-mobiles rates are > roughly half that and apparently do pass-thru charges for calls which don't > leave a given providers network...? Anyway, I spent nearly a month in > Spain this spring and my cell phone was my only contact, for both voice and > many long hours of GPRS internet access, and the bill was only $890 or > something similar. > > (I had a few 2.5k phone bills on similar length trips to England while > using AT&T...) that's useful information. --- > My daughter will take that Motorola V600! (She nags me daily to buy > one for her.) can't be done. i'm sending it back to AT&T Wireless as proof that i retained no value from the contract they violated. --- > Regretfully, the only three nation-wide GSM providers in the US > were AT&T Wireless, Cingular, and T-Mobile. AT&T Wireless recently > got bought by Cingular, so now there are only two nation-wide GSM > carriers in the US. Rumour has it that t-mobile is sharing > infrastructure with Cingular, so this would give you a serious SPOF to > consider. i don't worry too much about SPOF in this context. when the going gets tough, all cell phones stop working. > In fact, AT&T is fighting legal battles with Cingular over what > they have done to the AT&T Wireless brand since the sale, and > accelerating the timetable by which Cingular must switch over all > branding to their own name. This will allow AT&T to start using the > AT&T Wireless brand again, and resell CDMA coverage from one of the > other carriers. this mystifies me. i was an AT&T Wireless customer before may 19, and i've seen no evidence that my billing or customer service has yet been switched over to cingular. > ... And not all GSM carriers have roaming provisions with all other > GSM carriers -- I've been in the US in places where I got excellent > signal strength with the phones I've bought over here, but where I > couldn't get access to the network because there was no roaming > agreement with my carrier. i guess companies can believe that they're "tier 1" and that others should pay for access to their networks, even if the protocol isn't IP. > There's a good page comparing carriers in the SF Bay area at > <http://nordicgroup.us/ssub/sfbaratings.htm>, linked from > <http://sfbacell.com/>. ... that web page is 8 months old. a lot has changed since then. --- > If you haven't already, head over to www.howardforums.com and have a look > at their provider forums. One exists for every major cell phone provider. that's useful information. --- > Maybe this will help you Vixie.... > > http://www.fatwallet.com/forums/messageview.php?catid=18&threadid=348849 > &highlight_key=y&keyword1=at%26t not that i needed any additional evidence to show that at&t violated their contract with me, but this web site is useful and darkly humourous. to wit: Here's how - I've done it - so I know it works - you might have to call in a supervisor and fight it out - but legally you can cancel From August 18th AT&T Wireless is increasing their call charge for 411 calls to $1.50 from the current $1.25. The July bill has this mentioned as notice. This is considered 'Change of Service Terms' - and here's how the AT&T service contract refers to it: Changes to Agreement or Service. We may amend the terms of this Agreement, including the Sales Information, upon advance notice provided to you in any manner we choose, including by notice contained with your invoice for the Service. In the event that we make such a change that has a material adverse impact on your rights or use of the Service, you may terminate the Agreement by giving us notice within 20 days of the date we notify you, and you will not be charged any cancellation fee. If you use the Service more than 20 days after we notify you of a change, you agree to that change. Stick on to the fact that you received the July invoice in mail only around the 25th of the month - that'll keep you within the 20 allowable days to conflict this change in service terms All the best with your way out of the most horrible customer service in the world. the moral of this story: "in a race to the bottom, everyone loses." --- > Hey Paul, I just replaced my AT&T issued mMode phone with the following: > > http://www.resursadefun.ro/imagini/1handsfree_cellular.jpg > > Performance has been great. :-). --- > * Roaming with T-mobile is ~50% cheaper than roaming with ATT. This ratio > seems to be valid for most services. National calls, international calls > and GPRS roaming. (source: www.teliamobile.se) > > Overall, T-mobile seems to have better customer service, better > information on their homepage and better prices. Coverage-wise, they still > seem to be in the lead, although ATT is catching up. Telia's had roaming > with VoceStream/T-mobile for several years, but the ATT roaming agreement > was put in effect only 1-2 years ago. huh. i wonder how much it would cost me to become a telia customer and just roam when i'm in california ("mostly"). it's worth checking into. --- > -|> besides which, i hated the phone. i couldn't get it out of my > -|> pocket without hitting the voice-call button. the asynchronous > -|> nature of the java-based UI meant that the softkeys often changed > -|> what they meant while i was trying to press one. what a total > -|> piece of garbage. > -| > -|<CURIOSITY> So what model was this? </CURIOSITY> > > Actually, I missed that info as well. Could you pass this along to me > as well? I am about to go GSM shopping & want to know what phone to > avoid. Thanks. motorola V600. but any java-enabled phone will be asynchronous in this way. the rule is, if you're going to press a soft key, keep your eyes on the screen and be prepared to abort your button press if the screen changes. this happens most often when another call is coming in, or when the other call that's been trying to come in, stops trying. i'm expecting the NEC-515 to behave in exactly the same way, since UI these days is design-by-committee (or design-by-expert) and tends to trade common sense for glitz/gloss. but at least the NEC-515 should be able to come out of my pocket without being stuck in "dial by voice" mode. --- and that's all i know, for now. i've been without a cell phone for a week, and so far it's been really inconvenient for everybody else. if i don't get myself organized soon and get a replacement phone, i could really start to enjoy being unreachable when i'm not at my desk. (which is proof that there is no conspiracy.) === (i'm very sure that the nanog-powers-that-be would like this thread to end.) ***

1 0

MPLS signaling
by Paul Khavkine 08 Sep '04

08 Sep '04

Hi folks. What is the most common way of doing MPLS signaling for LSP setup and End to End QoS enforcement ? LDP ? RSVP ? BGP-TE or OSPF-TE ? Thanx Paul ************************************************ Paul Khavkine Network Administrator DISTRIBUTEL Communications. 740 Notre Dame West, Suite 1135 Montreal, Quebec, Canada, H3C 3X6 1-514-877-5505 x 263 http://www.distributel.net ************************************************

2 3

Re: Spammers Skirt IP Authentication Attempts
by Paul Jakma 08 Sep '04

08 Sep '04

On Wed, 8 Sep 2004, David Cantrell wrote: > Sure. But then, SPF is not meant to prevent all spam. s/all// > It would (if widely deployed) have a very wide effect on spam with > forged senders though, and the consequent bounces to innocent third > parties. I agree bounce reduction would be a good thing. But, I'm sure I've mentioned it already, *SRS achieves same thing* without breakage. To put it as an analogy (seems the done thing), I can use a plank of wood (SPF) to bang nails into a wall, but given the existence of hammer (SRS), the plank doesnt cut it ;) (and handing out planks so other people can make shelves isnt interesting). > The hype does not come from the people working on SPF, but from > clueless hangers-on who apparently can't read. Possible ;) Anyway, enough. The topic is indeed OT, and i'm just a net.kook. regards, -- Paul Jakma paul(a)clubi.ie paul(a)jakma.org Key ID: 64A2FF6A Fortune: Publishing a volume of verse is like dropping a rose petal down the Grand Canyon and waiting for the echo.

1 0

Re: Spammers Skirt IP Authentication Attempts
by Paul Jakma 08 Sep '04

08 Sep '04

On Wed, 8 Sep 2004, Florian Weimer wrote: > If there's significant deployment on the MTA side, it might > mitigate the impact of delayed bounces for those who post > authoritative SPF records. No offence, but did you read my reply to Paul Vixie? Delayed bounces, any any other NULL return path bot-mail, can be taken care of with SRS. > care about responsible Internet Mail handling. Therefore, it's > unlikely that we'll see significant deployment of a technology that > is as controversial as SPF. Good thing too. regards, -- Paul Jakma paul(a)clubi.ie paul(a)jakma.org Key ID: 64A2FF6A Fortune: You will be surprised by a loud noise.

1 0

Re:
by ken lindahl 08 Sep '04

08 Sep '04

At 10:21 AM 9/3/2004, Deepak Jain wrote: >>http://www.cisco.com/en/US/products/hw/modules/ps4835/ps5136/index.html >>But if there is no other traffic, why would you need deep packet buffers to beat records? (It doesn't say if this was on a production network with other traffic or not, so I don't know) > >I wouldn't suspect its on a production network at all. :) we might disagree wrt the definition of "production" :) http://ultralight.caltech.edu/lsr_06252004/ shows the path between caltech and cern transiting - a GSR in CENIC's LAX POP -- this is a production router for many higher ed institutions in california, including all university of california campuses, stanford, USC, caltech, and the california state university system. - abilene, the internet2 backbone, which is a production network for many higher ed institutions in the US. the abilene "weather map" image shown at the URL above shows that indeed there was other traffic on some of the links transited. ken > >DJ

2 1