HECnet January 2013

hecnet@lists.dfupdate.se

40 participants
1313 discussions

MOP vs. Cisco, was Re: [HECnet] _PROVISIONAL_ map of HECnet, courtesy largely of Brian H.

by mcguire＠neurotica.com

On 01/09/2013 01:57 PM, Johnny Billquist wrote: ...and like this from NCP under VMS: NCP> connect node gw physical address <MAC address> via <circuit-name> Note that the MAC address must have its octets delimited by colons under Linux, and hyphens under VMS. NCP>connect node gw physical address AA-00-04-00-01-F4 via ISA-0 Console connected (press CTRL/D when finished) User Access Verification Username: NCP> Uhhh...HOLY CRAP! That works from outside my network?! Is it working because you specified the circuit which gave it a path, or because it gleaned the DECnet area/node address from the (changed-by-DECnet) MAC address? I'm guessing the latter. (Peter?) It works on any LAN, including across bridges, so the various Internet based bridges we have make this work over some distance. The communication is via MOP remote console packets, which are not routed (just like LAT). The addressing is direct to the MAC address, whatever that happens to be at the destination. If the node you want to talk to supports MAC address aliases, the original address might work, but in any case the DECnet style Ethernet address will work. That was my belief before five minutes ago, when Brian was messing with me and got me all excited. I was certain that MOP was an unroutable, LAN-only protocol like LAT. But then, apparently folks have been booting DECservers via HECnet..? Nothing strange with that. The bridge makes it appear as if you are on the same ethernet segment. That makes sense...This will only work over bridged links, though. My Cisco GRE link (for example, in my configuration) should not propagate that traffic. (and if it does, is anyone else seeing my LAT service announcements? ;)) If (say) Brian and I were to create bridge-groups in our IOS configs that bridge across our Ethernet segments and the tunnel interfaces, that should work. I'm not too interested in doing that (it'd be lots of traffic) but I'd expect it to work. -Dave -- Dave McGuire, AK4HZ New Kensington, PA

11 years, 10 months

HECnet performance....

by bqt＠softjar.se

On 2013-01-09 19:33, Dave McGuire wrote: On 01/09/2013 12:11 PM, Cory Smelosky wrote: I would like to identify what results in me having rather low FAL access speeds, however. I don't know if my network is misconfigured, or if it's just a result of using real hardware on a remote node. Nearly all of the real hardware that I'm aware of on HECnet is actually very fast; significantly faster than pretty much any emulated system. The slowest system I'm aware of on here is my PDP-11/53. Keep in mind, on here you've got 600MHz Alphas (performance like a 2.5GHz x86 box but with better I/O) and you've got emulated VAXen running on 700MHz ARM processors...which do you think will be faster? ;) Still, really abmysal network file transfer speeds can happen for reasons that can be improved. Johnny

11 years, 10 months

HECnet performance....

by bqt＠softjar.se

On 2013-01-09 19:20, Peter Lothberg wrote: Also, the metrics are set on the circuit, and not individual destinations. While this worked more or less ok for DEC back in the day, with the bridge, the cost of two different destinations over the same circuit could in reality be very different. Bridged ethernet is a BAD thing for most protocols. And if you use spanning tree, it's not making optimal use of the topology. Totally agree. Radia finally came around and "Trill" is a routed network to carry ethernet frames. Hmm. Interesting. Might be something to look at, if I ever get plenty of time... As the cost/hop is in the routing vectors being sent periodiacally, you could implement a "cost/hop" NAT (half nat, one half on send one half on receive and some protocol to exchange how you want the metric matrix to look). That could be interesting, but since I don't expect us to fool around with the DECnet protocols as such, I guess this will not happen. Using dedicated routers that can tunnel traffic is really the best solution. Be it Ciscos, Multinet on VMS, or Rob's user level router... I think we're slowly getting more in that direction. Johnny

11 years, 10 months

HECnet performance....

by mcguire＠neurotica.com

On 01/09/2013 01:54 PM, Johnny Billquist wrote: On 2013-01-09 18:20, Brian Hechinger wrote: On 1/9/2013 12:15 PM, Jordi Guillaumes i Pons wrote: McLaughlin <ian at platinum.net> va escriure: I'd just like to say Thank You to everyone on the network for giving us all this huge plaything that we all dreamed of playing with a quarter of a century ago :) I strongly second this!! The awesomeness of all of you is beyond my ability to express in English! Mine too, and english is my native language! :) :-) And it all started just because I wanted to easily transfer data between my PDP-11 and home (a predecessor to PONDUS::), and the one at Update (MAGICA::). :-) And once it was working, why not let more people connect. Things are always more fun when shared. Absolutely agreed. Thank you Johnny! -Dave -- Dave McGuire, AK4HZ New Kensington, PA

11 years, 10 months

MOP vs. Cisco, was Re: [HECnet] _PROVISIONAL_ map of HECnet, courtesy largely of Brian H.

by bqt＠softjar.se

On 2013-01-09 18:58, Dave McGuire wrote: On 01/09/2013 12:54 PM, Paul_Koning at Dell.com wrote: ...and like this from NCP under VMS: NCP> connect node gw physical address <MAC address> via <circuit-name> Note that the MAC address must have its octets delimited by colons under Linux, and hyphens under VMS. NCP>connect node gw physical address AA-00-04-00-01-F4 via ISA-0 Console connected (press CTRL/D when finished) User Access Verification Username: NCP> Uhhh...HOLY CRAP! That works from outside my network?! Is it working because you specified the circuit which gave it a path, or because it gleaned the DECnet area/node address from the (changed-by-DECnet) MAC address? I'm guessing the latter. (Peter?) It works on any LAN, including across bridges, so the various Internet based bridges we have make this work over some distance. The communication is via MOP remote console packets, which are not routed (just like LAT). The addressing is direct to the MAC address, whatever that happens to be at the destination. If the node you want to talk to supports MAC address aliases, the original address might work, but in any case the DECnet style Ethernet address will work. That was my belief before five minutes ago, when Brian was messing with me and got me all excited. I was certain that MOP was an unroutable, LAN-only protocol like LAT. But then, apparently folks have been booting DECservers via HECnet..? Nothing strange with that. The bridge makes it appear as if you are on the same ethernet segment. Johnny

11 years, 10 months

HECnet performance....

by bqt＠softjar.se

On 2013-01-09 18:20, Brian Hechinger wrote: On 1/9/2013 12:15 PM, Jordi Guillaumes i Pons wrote: McLaughlin <ian at platinum.net> va escriure: I'd just like to say Thank You to everyone on the network for giving us all this huge plaything that we all dreamed of playing with a quarter of a century ago :) I strongly second this!! The awesomeness of all of you is beyond my ability to express in English! Mine too, and english is my native language! :) :-) And it all started just because I wanted to easily transfer data between my PDP-11 and home (a predecessor to PONDUS::), and the one at Update (MAGICA::). And once it was working, why not let more people connect. Things are always more fun when shared. Johnny

11 years, 10 months

MOP vs. Cisco, was Re: [HECnet] _PROVISIONAL_ map of HECnet, courtesy largely of Brian H.

by b4＠gewt.net

On 9 Jan 2013, at 13:11, Brian Hechinger <wonko at 4amlunch.net> wrote: On 1/9/2013 1:08 PM, Cory Smelosky wrote: On 9 Jan 2013, at 13:06, Brian Hechinger <wonko at 4amlunch.net> wrote: On 1/9/2013 1:03 PM, Dave McGuire wrote: On 01/09/2013 12:59 PM, Cory Smelosky wrote: I want in on your crazy GRE and DECnet routing setups! ;) I think of them as quite sane. If one needs to route, one should, oh you know, use a router. ;) I could install vyatta which is linux-based meaning I could /maybe/ build linux-decnet and have it function as an area router. Management could also be done via SNMP, but not to the ease of Cisco. Thoughts? Get a Cisco. It's not like they're tough to find, or very expensive. And you'll use it essentially forever. (because it will last that long!) I've got this 4700 sitting here collecting dust. Now that I have the 1841, 1811w and 2851 I seriously doubt I'll ever touch the 4700 again. Come get it. :) Where would I need to go? ;) Easton, PA. If you aren't familiar with the area, about half way betten NYC and Philly. Dave's gonna be heading my way soon, perhaps he could pick it up and bring it here to me? It'd be easier than me going there. -brian

11 years, 10 months

HECnet performance....

by roll＠Stupi.SE

On 1/9/2013 12:15 PM, Jordi Guillaumes i Pons wrote: McLaughlin <ian at platinum.net> va escriure: I'd just like to say Thank You to everyone on the network for giving us all this huge plaything that we all dreamed of playing with a quarter of a century ago :) I strongly second this!! The awesomeness of all of you is beyond my ability to express in English! Mine too, and english is my native language! :) And mine to ***THANKS*** ... And sorry for commanding some changes on other peoples domains, I just wanted some baseline functionality and a second order redundancy that would kick in by itself if things starts to break.. --P Any volounteers to write a portable NML layer and any any ideas on how to do a "show routing table" using the standard NCP stuff?

11 years, 10 months

MOP vs. Cisco, was Re: [HECnet] _PROVISIONAL_ map of HECnet, courtesy largely of Brian H.

by ian＠platinum.net

On 2013-01-09, at 10:17 AM, Cory Smelosky <b4 at gewt.net> wrote: Ooooh. It looks interesting. Now I just need to find an IOS image. ;) Just sent you a couple of Dropbox links to some files you may find useful... Ian

11 years, 10 months

HECnet performance....

by mcguire＠neurotica.com

On 01/09/2013 12:11 PM, Cory Smelosky wrote: I would like to identify what results in me having rather low FAL access speeds, however. I don't know if my network is misconfigured, or if it's just a result of using real hardware on a remote node. Nearly all of the real hardware that I'm aware of on HECnet is actually very fast; significantly faster than pretty much any emulated system. The slowest system I'm aware of on here is my PDP-11/53. Keep in mind, on here you've got 600MHz Alphas (performance like a 2.5GHz x86 box but with better I/O) and you've got emulated VAXen running on 700MHz ARM processors...which do you think will be faster? ;) -Dave -- Dave McGuire, AK4HZ New Kensington, PA

11 years, 10 months

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

HECnet January 2013