Re: Network performance 6.0 with netperf
At 10:49 PM +1000 2005-10-20, Michael VInce wrote: The 4 ethernet ports on the Dell server are all built-in so I am assuming they are on the best bus available. In my experience, the terms "Dell" and "best available" very rarely go together. Dell has made a name for themselves by shipping the absolutely cheapest possible hardware they can, with the thinnest possible profit margins, and trying to make up the difference in volume. Issues like support, ease of management, freedom from overheating, etc... get secondary or tertiary consideration, if they get any consideration at all. But maybe that's just me. -- Brad Knowles, <[EMAIL PROTECTED]> "Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety." -- Benjamin Franklin (1706-1790), reply of the Pennsylvania Assembly to the Governor, November 11, 1755 SAGE member since 1995. See <http://www.sage.org/> for more info. ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Network performance 6.0 with netperf
At 9:57 AM -0500 2005-10-20, Karl Denninger wrote: Other than that, I've been pretty happy with their stuff. Sure beats a lot of other "PC" vendors out there in terms of reliability, heat management, BIOS updates, etc. Have you tried Rackable or IronSystems? I've heard that they've been pretty successful at building servers to compete pretty well on price with Dell, while also providing much better customer service, including custom-building servers to your precise requirements. -- Brad Knowles, <[EMAIL PROTECTED]> "Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety." -- Benjamin Franklin (1706-1790), reply of the Pennsylvania Assembly to the Governor, November 11, 1755 SAGE member since 1995. See <http://www.sage.org/> for more info. ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: My planned work on networking stack
At 11:26 AM +0300 2004/03/02, Gleb Smirnoff wrote: Is there any plans about integration of BGP routing daemon (Zebra or Quagga) into FreeBSD? With BGP routing daemon onboard, FreeBSD will be a strong alternative against expensive commercial routers. I have successfull experience of running FreeBSD STABLE with 2 full BGP views for half a year. Modern i386 PC can route/filter/shape much more traffic than expensive Cisco 36xx. I haven't yet compared with 7000 series... Talk to people who have real-world experience in running zebra/quagga in ISP environments with multiple upstreams and taking full views. The guy who is designing bgpd for OpenBSD gave a talk on the subject at FOSDEM, and it was very enlightening to hear about the problems with zebra (which went commercial and the open source version basically hasn't been touched in years) and quagga (which is a community of zebra users trying desperately to fix the worst of the bugs), and how he has used this information during his design of a replacement, and the methodology he used to make sure that the resulting system is robust and capable of being used in real-world production environments. His only issue with using exclusively PC equipment for handling routing is all those strange WAN protocols and cards for which hardware cards are rarely available beyond vendors like cisco or Juniper. That's why he's going pure Ethernet protocols/hardware throughout all his networks, including his upstream feeds, so that he can dump all that expensive ancient legacy routing hardware. If anything, I'd be inclined to look towards his work for OpenBSD and see if that could be imported into FreeBSD (and maybe improved, with contributions given back to him), rather than mess around with crap like zebra or quagga. Oh, and it would be nice if someone somewhere started thinking about a mesh routing implementation for *BSD, either AODV or something else. -- Brad Knowles, <[EMAIL PROTECTED]> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+() DI+() D+(++) G+() e++> h--- r---(+++)* z(+++) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: My planned work on networking stack
At 3:59 PM +0300 2004/03/02, Gleb Smirnoff wrote: Haven't you understand? I'm the "person who has real-world experience in running zebra in ISP environments with multiple upstreams and taking full views". Do you have multiple connectivity to two separate metro area exchanges, with multiple upstreams at each? Most large cities are lucky to have a single major metro area exchange, and the author of bgpd for OpenBSD works at an ISP located in Hamburg which is lucky enough to have two major NAPs, and he has multiple connectivity to both. He was the one ragging on zebra/quagga. Among other things, he said he had real problems keeping sessions up with zebra/quagga when neighbors were flapping. IIRC, he's also got some pretty big cisco equipment (75xx or whatever), and he is going to be switching over to OpenBSD+bgpd as his secondary core router in the very near future, with plans to complete the switch over soon thereafter. He's putting his money where his mouth is. Certainly, I have noticed that zebra hasn't done much recently, and at least on the surface quagga doesn't seem to have gone that far beyond where zebra was a couple of years ago. Browse zebra CVS to make sure that author is commiting bugfixes. For example: last commit to BGP code is done 2 weeks ago. Right, and that bugfix took how long to apply? When was the previous bugfix before that? When was the last real "new" development for zebra? I can't say a word about quagga, since I haven't use it, but I have positive experience with zebra (see above). If you're a zebra fan, then I suggest you check out quagga. I stop replying... Do not like flame. Before flaming anyone further, you might want to check out pages like <http://www.fosdem.org/2004/index/interviews/interviews_brauer>, and then take a look and see what Henning Brauer has actually been up to. You might also want to check out <http://www.commsdesign.com/design_corner/OEG20010323S0048> and ask yourself if zebra/quagga handles resiliency the way it should. If this problem isn't already addressed by bgpd, I'm sure it will be before Henning can go production with using this for his core routers at his ISP. -- Brad Knowles, <[EMAIL PROTECTED]> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+() DI+() D+(++) G+() e++> h--- r---(+++)* z(+++) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: My planned work on networking stack
At 3:59 PM +0100 2004/03/02, Andre Oppermann wrote: Zebra is definatly *not* a piece of s*** as you make it sound here. Well, that was certainly the impression I got from Henning Brauer at FOSDEM. Maybe I misunderstood him, or maybe he has different views on this software than you do. To properly clarify this matter, you should probably talk to him directly and find out if his opinions on zebra/quagga really are as different from yours, and if so then I'll be glad to make a public apology. However, until then, I will stand by what I remember of his talk, and neither you nor anyone else is going to make me change my mind, certainly not by beating your chest. You need GigE, T1/E1, E3/T3 and STM-1 these days. Everything else is dead. From what I understand from Henning, he's going to be dumping E-1/T-1, E3-T3, and probably also STM-1, because you can't get those kinds of interfaces for regular PC-type boxes. I'm not sure I agree with him 100%, but I can certainly understand why he'd want to simplify his life. Ok, again Zebra/Quagga is not "crap". The same with DJBware which is no "crap" either. If you don't like it just say so but refrain from dirt-talking it. It doesn't make your point any stronger. Beating your chest louder is not likely to make me believe that you're right. If you want to get off onto a "Church of Dan" rant, I can certainly do that, and I can point out a whole ark-load of flaws -- most of which are simple basic facts which Dan himself admits to, but when he says them they're "facts" and when I say them they're "libel" or "slander". Yeah, riight. The bgpd from OpenBSD will surely make it's way into FreeBSD [*]. The main developer besides Henning sits about 5 meters away from me in my office. If you look at it then you'll find out that I'm not really innocent that bgpd ;-) I'm glad to hear that. [*] In FreeBSD it will be a port. I don't know why a bgpd should be in the base system. I don't know. Why should we have any routing software at all in the base system? It would be nice if you could calm down, stop your mis-informed accusations and rants and actually try to be helpful and progressive to the projects which try to do it better. Thank you very much. Show me the words from Henning himself where I have mis-represented his views on zebra/quagga, and I will gladly apologize in public. Until then, I stand by what I have said. -- Brad Knowles, <[EMAIL PROTECTED]> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+() DI+() D+(++) G+() e++> h--- r---(+++)* z(+++) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: My planned work on networking stack
At 4:08 PM +0100 2004/03/02, Andre Oppermann wrote: Gleb is doing the same, and so am I. However you are not. Do you run BGP in your network? I'm not running an ISP that is multiply connected to at least two metro-area NAPs and has multiple upstreams at both sites, no. I would be very interested to be involved in the network management of a medium to large-sized ISP, however. At least for me on FreeBSD Zebra has been very stable for me. There is no need to always "change" things. That's wonderful for you. However, that doesn't change the criticism that Henning has levelled at zebra/quagga. What is you point? Do you use Zebra? Are you affected by it? Or are you just ranting? My point is that zebra/quagga have significant limitations that restrict their usefulness, due to the design of the system. Moreover, the development on zebra has effectively stalled since the author got hired away to do that kind of work professionally, and development on quagga has apparently been sporadic and relatively limited, presumably due to the fact that they don't have replacement developers of the same caliber. If we want to get to the point where we can have a reasonable expectation of throwing away all cisco, juniper, Foundry, and other routing hardware and replace them with something that is easier to install, configure, monitor, and manage, then I think we need to be looking beyond zebra/quagga. And you should stop flaming anyone if you haven't ever used or done what you are blabbering about. If you think this is flaming, then you have never seen flaming. At this stage, this is nothing more than a luke-warm disagreement. Sorry, but OpenBSDs bgpd wont to any of that either. This is mostly hardware that needs to be redundant. Not much you can in bgpd. Not in bgpd per se, no. But by then you'd have added more protocol support to the daemon and that name would no longer be appropriate. -- Brad Knowles, <[EMAIL PROTECTED]> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+() DI+() D+(++) G+() e++> h--- r---(+++)* z(+++) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: My planned work on networking stack
At 11:02 AM +0200 2004/03/02, Andrew Degtiariov wrote: What's difference (*currently*) beetwen FreeBSD+Zebra and Cisco routers? Support for VRRP? Support for various other routing protocols not covered by zebra/quagga -- at least not yet, if ever? Support for line cards and other devices that do not exist in a format you can plug into a PC? Maybe there's nothing you can do about this last item, but there's plenty that can be done on the software side -- just take a look at all the protocols that have been identified as being desirable, but not yet implemented by zebra/quagga. Oh, and then there are all the operational issues where zebra/quagga can't keep sessions going when a neighbor flaps, etc Those would require re-architecting the whole routing system, at which point it might make a lot more sense to go with a different implementation -- such as bgpd from OpenBSD. -- Brad Knowles, <[EMAIL PROTECTED]> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+() DI+() D+(++) G+() e++> h--- r---(+++)* z(+++) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: My planned work on networking stack
At 3:52 PM +0200 2004/03/02, Andrew Degtiariov wrote: Oh, and then there are all the operational issues where zebra/quagga can't keep sessions going when a neighbor flaps, etc Those would require re-architecting the whole routing system, at ^^^ Congratulation. That's namely what the conversation was about. Right. We can either re-architect zebra/quagga, or we can start with something that addresses the weaknesses in these tools, or we can do something else. I'm advocating that we at least take a long hard look at what Henning Brauer has done, and seriously consider whether it would make sense for us to start with that to give us a leg up on the re-architecting process. If nothing else, this would at least give us an interesting insight to what some of the weaknesses are in this category, and maybe help us identify better solutions faster and more easily. In particular, if there are such serious problems with zebra/quagga that they would need to be completely re-architected in order to be useful, then I don't see that as being a particularly fruitful line of work to pursue. I'd rather start with something that requires less re-work, and would presumably allow us to more easily add in any additional bits that we feel are necessary or desirable. -- Brad Knowles, <[EMAIL PROTECTED]> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+() DI+() D+(++) G+() e++> h--- r---(+++)* z(+++) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: My planned work on networking stack
At 4:18 PM +0100 2004/03/02, Andre Oppermann wrote: I'd like to see you do any real work in this area instead of producing many and longs emails with lots of mis-informed rants in them. Yes, this my official put-up-or-shut-up call to you. I'm not a programmer. I haven't done anything that I consider to be proper "programming" in over fifteen years. If there is anything I can do to help with the skills I have as a senior unix systems administrator and a small network of machines downstairs that I need to put together (four UltraSPARC 10 clones, a dishwasher-size four-processor Intel OEM fileserver-to-be, an ancient SPARC-4 clone, and an ancient Pentium-133 laptop w/ 48MB of RAM), then I'll be glad to do what I can to help. -- Brad Knowles, <[EMAIL PROTECTED]> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+() DI+() D+(++) G+() e++> h--- r---(+++)* z(+++) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Who wants SACK? (Re: was My planned work on networking stack)
At 3:32 PM -0800 2004/03/08, Jeffrey Hsu wrote: What Luigi says is absolutely correct. It doesn't take a lot to get this done. I've talked to a number of companies about implementing SACK for them and while there was interest, no one wanted to fund it all themselves, potentially for the benefit of their competitors. Out of curiosity, can someone provide some pointers as to where SACK really helps? Is this just for high-speed WANs and doesn't help on LANs, or is it useful in both contexts? Also, at what speeds/packet sizes does SACK start to become really useful? I'm just wondering if there aren't a lot of people who could benefit from something like this, only they don't know it. If they were to find out, it might help provide funding and other resources to spur development. -- Brad Knowles, <[EMAIL PROTECTED]> "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." -Benjamin Franklin, Historical Review of Pennsylvania. GCS/IT d+(-) s:+(++)>: a C++(+++)$ UMBSHI$ P+>++ L+ !E-(---) W+++(--) N+ !w--- O- M++ V PS++(+++) PE- Y+(++) PGP>+++ t+(+++) 5++(+++) X++(+++) R+(+++) tv+(+++) b+() DI+() D+(++) G+() e++> h--- r---(+++)* z(+++) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net To unsubscribe, send any mail to "[EMAIL PROTECTED]"
