- Test - lists.nanog.org

Re: GigaRouter (Was Re: Cisco as Big Brother))
by Rob Liebschutz 21 Oct '96

21 Oct '96

> Michael Dillon writes: > > You can also try building a machine with a boot device like the 2.88 > > megabyte floppies. Using the same techniques FreeBSD uses for their boot > > disks, you can decompress the boot floppy into a large RAMDISK and run > > that way. Or simply use a ZIP drive for the boot device but run from RAM > > as before. It's not as good as 100% solid state but it comes pretty close. > > This isn't clear to me. Why do you assume a ZIP is likely to be more reliable > that a hard disk? ZIPs haven't been around long enough to be sure of this, > and HDs are pretty reliable these days. > > Of course I'm not saying that I *Want* to use an HD in this situation; flash > is clearly a big win. But I don't see how using a floppy or ZIP improves > wins. > > FWIW, I suspect that building a 1.4MB fs that can boot and then nfs-mount > (or ftp to a memory fs) needed binaries would be not a lot harder for FreeBSD > or BSDi than it was for NetBSD. Not really, the BSDI installation procedure does it with one floppy. The BSDI bootstrap knows how to load a gzipped kernel image. I've made modified boot disks from their installation procedure. I've never turned on NFS though, and admittedly it's all packed in there tight enough that every time I want to do something else I'm deleting one thing to add another. An 8mb flash card for $300 would be very nice if it worked. Rob

2 1

RE: Windows/NT to break the Internet?
by Hank Nussbacher 21 Oct '96

21 Oct '96

On Fri, 18 Oct 1996 11:16:28 -0700 Peter Ford wrote: > >Hank, > >Have your tried to replicate your customer's problem? We have tested >setting the prefix mask in the IP stack in NT 4.0 and it can be set to >prefixes longer than 24. > >cheers, peter I have to apologize for the lack of clarity since I was getting the reports from customers. The problem is NOT in setting a prefix mask. The problem is with inverse for subnets. From the install notes of Windows/NT (DNS and Microsoft Windows NT 4.0) dated May 12, 1996 (page 28), under the section 'Configuring Reverse Lookup' the doc says, and I quote: ??? I haven't a clue. I think there is a bug??? end quote. The defacto standard for doing subnetting and inverse has been documented in a draft standard: ftp://ietf.cnri.reston.va.us/internet-drafts/draft-ietf-cidrd-classless-inaddr-00.txt I hope Microsoft can get this fixed since I have had to start allocating /24s to very small sites in order to bypass this problem. Hank

1 0

No. of routers carrying full routes?
by dlc＠silcom.com 21 Oct '96

21 Oct '96

Where can I find/how can one estimate the number of routers carrying the full Internet route tables? I need to convince someone that singly-homed customer route flaps/ withdrawals should *not* propagate beyond our AS. I've found some discussion of this in the July NANOG archives, and talk about cisco floating statics, etc... and that "one-way" traffic is insignificant. I haven't upgraded to 11.1 yet, but my network is simple enough [read: not redundant enough?] that all my customers are singly-connected; in that case, statics in my border routers pointing to my customer edge-router seems to make fast install moot for now. Finally, is there a *searchable* archive of NANOG anywhere? Thanks. -- David Carmean WB6YZM DC574 <dlc(a)silcom.com> System/Network Administration, Silicon Beach Communications Unsolicited commercial e-mail not accepted. Violators will be LARTed.

2 2

Re: GigaRouter (Was Re: Cisco as Big Brother))
by Alexis Rosen 21 Oct '96

21 Oct '96

Neil J. McRae writes: > Well if you're talking i386. For NetBSD/i386 work is being done to > make gzip'ed kernels boot at thi time. So the author told me. Thing is, he's as bad in some ways as I am. Maybe it'll be done in two weeks, maybe not. But whenever it's done, I'm sure it'll be great. BTW, I'm curious why some people don't seem to like the idea. At the speed of modern systems, I suspect that loading uncompressed off flopy isn't much faster (if at all) than loading compressed off floppy and decompressing. In fact it may be slower. I'd test it right now but my arms aren't 4 miles long, which they'd have to be to reach the closest floppy drive on one of my NetBSDs... Heh. I can just see it. gzip becomes a kernel function. Although that's not really a joke- compressing filesystems aren't far off that mark... /a

2 1

Re: GigaRouter Latency?
by Vadim Antonov 21 Oct '96

21 Oct '96

Deepak Jain <deepak(a)jain.com> wrote: >Latency in a router is only somewhat important for most network >applications but on some applications it has a very serious effect. If >anyone would like to flame me, please do it in private email and I am >sure I will find the time to get back to you :). I'd like to point out that 1 ns of latency equivalent to 1 ft of wire (more-less), so 1 ms of latency is 190 miles of wire. It is not a serious delay, when compared with typical thousand-mile backbone links. (1 ms is a _lot_ of time to route a packet). --vadim

2 1

Re: GigaRouter (Was Re: Cisco as Big Brother))
by Vadim Antonov 20 Oct '96

20 Oct '96

I beieve my troubles with cisco flash were due to the relatively frequent updates (once a week or so) and somewhat high ambient temperatures (there wasn't adequate air flow). Solid-state nonvolatile memories age rather eapidly in those conditions. --vadim

1 0

Re: GigaRouter (Was Re: Cisco as Big Brother))
by Rob Liebschutz 20 Oct '96

20 Oct '96

> Alexis Rosen <alexis(a)panix.com> wrote: > > >Of course I'm not saying that I *Want* to use an HD in this situation; flash > >is clearly a big win. > > In my experience flash is a lot _less_ reliable than HDs. Ciscos losing > flash contents aren't that uncommon. Is this a general statement about flash, or just about the flash in Ciscos? I'd find it hard to believe that well designed solid state devices could be lest reliable than a disk drive, especially since a large part of a disk drive is solid state as well. > > >But I don't see how using a floppy or ZIP improves > >wins. > > Floppy doesn't rotate until it's boot time; and in COs dust is not > a big problem. > > So, for a boot device floppy is probably the best (providing it's not > an el cheapo device). An additional benefit is that it's easy for > field personnel to handle. > > --vadim >

2 1

Re: GigaRouter (Was Re: Cisco as Big Brother))
by Robert Sanders 20 Oct '96

20 Oct '96

Vadim Antonov <avg(a)quake.net> writes: > In my experience flash is a lot _less_ reliable than HDs. Ciscos losing > flash contents aren't that uncommon. ... > So, for a boot device floppy is probably the best (providing it's not In my experience, in an equal amount of hours of running Netblazers and Ciscos, the routers' flash has never failed me while the Netblazers *all* eventually failed to boot from floppy. The environment wasn't especially dusty, but it wasn't a clean room. I'd hesitate to use any floppy-booted device intended for use out of arm's reach. -- Robert

1 0

Re: GigaRouter (Was Re: Cisco as Big Brother))
by Rob Liebschutz 20 Oct '96

20 Oct '96

> > From: Michael Dillon <michael(a)memra.com> > > On Sat, 19 Oct 1996, Robert E. Seastrom wrote: > > > With cold-convenient-swappable IDE drawers that let even a > > kindergartener swap out an IDE hard drive and high quality 100mb hard > > drives available for like $50 > > Where on earth do you find new drives this small??? > > http://www.corpsys.com > > But apparently with their current stock of new drives you'll have to > settle for IBM 3.5" 170mb drives for $99 each ($89 qty 10). > > Now, if you don't mind refurbs, there are $65/$55 80mb drives and > $69/$59 131mb drives. I've dealt with these folks before and they're > righteous. > > Oh well, the $50 was a little optimistic, or maybe not if you're > buying a bunch of 'em... > > ---Rob > > > I have a hard time calling IDE drives reliable. At one time I used them in firewalls, and every now and then, tripwire would report random changed bits on the disk, and then the next time I ran tripwire, the problem would just go away. Now I use low end SCSI drives for these kind of applications cause I prefer to have all my data parity checked. Rob

1 0

GigaRouter Latency?
by Deepak Jain 19 Oct '96

19 Oct '96

Now I know a lot of people are going to jump on me for raising the latency question here, and I don't want to inflate its importance at all. Latency in a router is only somewhat important for most network applications but on some applications it has a very serious effect. If anyone would like to flame me, please do it in private email and I am sure I will find the time to get back to you :). How does the latency through a Gigarouter compare to the latency of say a NetBSD or FreeBSD box with (say) FDDI interfaces. [Ignoring the fact that most BSD boxes are running on a single PCI bus]. Basically I am curious how much advantage/disadvantage one gets by going from a Cisco to a Gigarouter solution based on issues like latency. I am sure the difference we are talking about is very slight and <2ms total, I just want to get an idea and I don't trust most of the technical sales personnel I have talked to to get a straight answer. I am very familiar with the latency of BSD (typical) boxes and those of Cisco Routers. Thanks, -Deepak.

2 1