Send Outages-discussion mailing list submissions to outages-discussion@outages.org
To subscribe or unsubscribe via the World Wide Web, visit https://puck.nether.net/mailman/listinfo/outages-discussion or, via email, send a message with subject or body 'help' to outages-discussion-requ...@outages.org You can reach the person managing the list at outages-discussion-ow...@outages.org When replying, please edit your Subject line so it is more specific than "Re: Contents of Outages-discussion digest..." Today's Topics: 1. Re: NTT - High Latency between Dallas and LA exchanges (Joseph Jackson) 2. Re: NTT - High Latency between Dallas and LA exchanges (Patrick W. Gilmore) 3. Re: NTT - High Latency between Dallas and LA exchanges (John Kristoff) 4. Re: NTT - High Latency between Dallas and LA exchanges (Ross Tajvar) ---------------------------------------------------------------------- Message: 1 Date: Thu, 18 May 2023 13:04:20 +0000 From: Joseph Jackson <jjack...@aninetworks.net> To: "Patrick W. Gilmore" <patr...@ianai.net>, Gert Doering <g...@greenie.muc.de> Cc: Outages Discussion <outages-discussion@outages.org> Subject: Re: [Outages-discussion] NTT - High Latency between Dallas and LA exchanges Message-ID: <cy8pr17mb62603456c72dfcb2871d2ddddb...@cy8pr17mb6260.namprd17.prod.outlook.com> Content-Type: text/plain; charset="iso-8859-1" As a voip provider whose traffic is almost all UDP and a lot of it I had no idea this was something that people, much less ISPs thought. I have never come across the idea that UDP traffic through routers at least in my experience was being rate limited. Joseph ________________________________ From: Outages-discussion <outages-discussion-boun...@outages.org> on behalf of Gert Doering via Outages-discussion <outages-discussion@outages.org> Sent: Thursday, May 18, 2023 7:56 AM To: Patrick W. Gilmore <patr...@ianai.net> Cc: Outages Discussion <outages-discussion@outages.org> Subject: Re: [Outages-discussion] NTT - High Latency between Dallas and LA exchanges Hi, On Thu, May 18, 2023 at 08:18:58AM -0400, Patrick W. Gilmore via Outages-discussion wrote: > Yet you think this is Google???s fault by advocating for an open protocol > which has objective benefits to end users? What am I missing? Building a new protocol on top of UDP when it's well-known that many ISPs rate-limit UDP (due to "there is no large amounts of UDP in the wide area Internet, except for reflective DDoS crap") is not exactly a very smart move. Nothing about QUIC is really a smart move, beyond "we're google, we can do what we want" - and IETF being what it is, if you have strong enough vendor backing, you can get anything standardized. gert -- "If was one thing all people took for granted, was conviction that if you feed honest figures into a computer, honest figures come out. Never doubted it myself till I met a computer with a sense of humor." Robert A. Heinlein, The Moon is a Harsh Mistress Gert Doering - Munich, Germany g...@greenie.muc.de _______________________________________________ Outages-discussion mailing list Outages-discussion@outages.org https://puck.nether.net/mailman/listinfo/outages-discussion -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://puck.nether.net/pipermail/outages-discussion/attachments/20230518/c340e210/attachment-0001.htm> ------------------------------ Message: 2 Date: Thu, 18 May 2023 09:09:10 -0400 From: "Patrick W. Gilmore" <patr...@ianai.net> To: Outages Discussion <outages-discussion@outages.org> Subject: Re: [Outages-discussion] NTT - High Latency between Dallas and LA exchanges Message-ID: <1826ce4b-e33e-4044-ac64-219cacc5c...@ianai.net> Content-Type: text/plain; charset="utf-8" On May 18, 2023, at 08:56, Gert Doering <g...@greenie.muc.de> wrote: > On Thu, May 18, 2023 at 08:18:58AM -0400, Patrick W. Gilmore via > Outages-discussion wrote: >> Yet you think this is Google???s fault by advocating for an open protocol >> which has objective benefits to end users? What am I missing? > > Building a new protocol on top of UDP when it's well-known that many > ISPs rate-limit UDP (due to "there is no large amounts of UDP in the > wide area Internet, except for reflective DDoS crap") is not exactly > a very smart move. > > Nothing about QUIC is really a smart move, beyond "we're google, we can > do what we want" - and IETF being what it is, if you have strong enough > vendor backing, you can get anything standardized. TCP has run into problems and limitations Dr. Kahn & Dr. Cerf could not have envisioned in the 70s. (I guess 1980 is when v4 was finally standardized, but still.) QUIC avoids some of those limitations and has helped improve the experience of literally billions of people (and devices). How exactly does that equate to not being ?a smart move?? As for ?vendor backing?, it takes more than an IETF RFC to be accepted and adopted. (Citation: IPv6.) Also, I was specifically speaking of "side-step the protocol stack?. It is, by definition, not side-stepping the protocol stack. Look, if you don?t like QUIC, no worries. I do not work for Google, and I am not trying to tell you how to run your apps or your network. But QUIC is a real protocol used by real people that has real benefits over TCP/HTTP. Claiming otherwise is, frankly, silly. -- TTFN, patrick -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://puck.nether.net/pipermail/outages-discussion/attachments/20230518/b00ed15d/attachment-0001.htm> ------------------------------ Message: 3 Date: Thu, 18 May 2023 08:59:00 -0500 From: John Kristoff <j...@dataplane.org> To: Joseph Jackson via Outages-discussion <outages-discussion@outages.org> Subject: Re: [Outages-discussion] NTT - High Latency between Dallas and LA exchanges Message-ID: <20230518085900.28747...@dataplane.org> Content-Type: text/plain; charset=US-ASCII On Thu, 18 May 2023 13:04:20 +0000 Joseph Jackson via Outages-discussion <outages-discussion@outages.org> wrote: > As a voip provider whose traffic is almost all UDP and a lot of it I > had no idea this was something that people, much less ISPs thought. > I have never come across the idea that UDP traffic through routers at > least in my experience was being rate limited. I don't know how widespread it is, but this was precisely something I had done many years ago before QUIC. Slammer was what stimulated me to implement "edge" rate limits on UDP traffic towards external destinations at an edu. So for example, ingress to the network traffic from an end user subnet I set a max of 10 Mb/s for UDP traffic not destined to internal prefixes. I left the organization and came back years later. When there were complaints of some random real-time game performance I discovered someone had later put an aggregate limit of about 100 to 200 Mb/s for UDP at peering routers, and with the rise of the QUIC, that limit was now being reached by the total sum of UDP traffic from all internal subnets. I preceded to get rid of the hard coded UDP limits with this new reality. It seemed like a reasonable thing to do at the time, but not so much now. Like manually configured bogon filters I would assume there may be similar cases lurking out there. John ------------------------------ Message: 4 Date: Thu, 18 May 2023 15:09:53 -0400 From: Ross Tajvar <r...@tajvar.io> To: John Kristoff <j...@dataplane.org> Cc: Joseph Jackson via Outages-discussion <outages-discussion@outages.org> Subject: Re: [Outages-discussion] NTT - High Latency between Dallas and LA exchanges Message-ID: <ca+fdddqhuvr+1ostkq7uplrqufw2xiq8tkmh65nzjsbwcug...@mail.gmail.com> Content-Type: text/plain; charset="utf-8" It makes sense to heavily rate-limit certain UDP traffic that "should" not be much on the DFZ and is commonly used in amplification attacks (things like SSDP, LDAP, memcached, etc.). NTT does this on all customer ports. Rate-limiting ALL UDP in 2023 is a very bad idea. On Thu, May 18, 2023 at 10:00?AM John Kristoff via Outages-discussion < outages-discussion@outages.org> wrote: > On Thu, 18 May 2023 13:04:20 +0000 > Joseph Jackson via Outages-discussion <outages-discussion@outages.org> > wrote: > > > As a voip provider whose traffic is almost all UDP and a lot of it I > > had no idea this was something that people, much less ISPs thought. > > I have never come across the idea that UDP traffic through routers at > > least in my experience was being rate limited. > > I don't know how widespread it is, but this was precisely something I > had done many years ago before QUIC. Slammer was what stimulated me to > implement "edge" rate limits on UDP traffic towards external > destinations at an edu. So for example, ingress to the network > traffic from an end user subnet I set a max of 10 Mb/s for UDP traffic > not destined to internal prefixes. > > I left the organization and came back years later. When there were > complaints of some random real-time game performance I discovered > someone had later put an aggregate limit of about 100 to 200 Mb/s for > UDP at peering routers, and with the rise of the QUIC, that limit was > now being reached by the total sum of UDP traffic from all internal > subnets. I preceded to get rid of the hard coded UDP limits with this > new reality. It seemed like a reasonable thing to do at the time, but > not so much now. Like manually configured bogon filters I would assume > there may be similar cases lurking out there. > > John > _______________________________________________ > Outages-discussion mailing list > Outages-discussion@outages.org > https://puck.nether.net/mailman/listinfo/outages-discussion > -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://puck.nether.net/pipermail/outages-discussion/attachments/20230518/39192b1b/attachment.htm> ------------------------------ Subject: Digest Footer _______________________________________________ Outages-discussion mailing list Outages-discussion@outages.org https://puck.nether.net/mailman/listinfo/outages-discussion ------------------------------ End of Outages-discussion Digest, Vol 155, Issue 4 **************************************************