HECnet

hecnet@lists.dfupdate.se

8 participants
15562 discussions

by wonko＠4amlunch.net

On 1/9/2013 11:44 AM, Johnny Billquist wrote: On 2013-01-09 17:19, Peter Lothberg wrote: On 1/9/2013 1:59 PM, Peter Lothberg wrote: On 1/8/2013 9:01 PM, Peter Lothberg wrote: Set up a (cisco) tunnel to 130.238.19.60 (and tell me your side IP address). I think it's time to come up with something a little better to track all these. Thoughts? Build a tool that takes requests and generates configs (full mesh) and pushes them to the routers. (maybe we can have the router configs in a wiki?) Great mind think alike. :) I had though pretty much exactly that. If we are going to do that, however, we should decide on a range of tunnels to use. For example on my router, 0, 1 and 2 are DECnet tunnels, 3 is my IPv6 tunnel to HE and 4 is a DECnet tunnel. Re-pushing the configs in this case would be ugly, so I think it would be wise to say, tunnels X through Y are dedicated to auto-DECnet tunnels so I can write cisco configs to deal with that appropriately. I will write something. :) -brian If you knew the capability (what kind of tunnels) a node can support and you have a fixed priority for in what order to chose, given two nodes you can automatically determine the required tunnel type and don't need to encode it in to you global description as it's a pair-wise local matter..... -:) A big part of the general problem is that all these numbers are local for each end of connections, so that you can get very asymetrical setups. 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. In the end, it boils down to a couple of things for tweaking the costs. 1. Preference of tunnel or bridge. If you prefer using the bridge, then just set the ethernet circuit cost lower than the tunnel. The tunnel can be either a multinet tunnel, or a tunnel on the cisco. 2. If the rest of your network isn't on the same ethernet as the bridge is, then you can use tweaks and control to your hearts content. 3. If the rest of your network don't have some other path out, then it don't matter if the ethernet circuit cost is high, so still feel free to tweak it as needed. I think that you should be able to tweak the costs freely almost all the time, without any issues. However, you will sometimes not get the results you are looking for if the other end, and all intermediate nodes have also set their costs based on similar views as yours. I was referring to the Cisco tunnel interface numbering, not path cost. You did address a question I had about that, however. :) I can make it configurable, but ultimately it might be easiest to choose a path cost that is most likely going to work for everyone. -brian -brian

12 years, 5 months

HECnet performance....

by bqt＠softjar.se

On 2013-01-09 17:19, Peter Lothberg wrote: On 1/9/2013 1:59 PM, Peter Lothberg wrote: On 1/8/2013 9:01 PM, Peter Lothberg wrote: Set up a (cisco) tunnel to 130.238.19.60 (and tell me your side IP address). I think it's time to come up with something a little better to track all these. Thoughts? Build a tool that takes requests and generates configs (full mesh) and pushes them to the routers. (maybe we can have the router configs in a wiki?) Great mind think alike. :) I had though pretty much exactly that. If we are going to do that, however, we should decide on a range of tunnels to use. For example on my router, 0, 1 and 2 are DECnet tunnels, 3 is my IPv6 tunnel to HE and 4 is a DECnet tunnel. Re-pushing the configs in this case would be ugly, so I think it would be wise to say, tunnels X through Y are dedicated to auto-DECnet tunnels so I can write cisco configs to deal with that appropriately. I will write something. :) -brian If you knew the capability (what kind of tunnels) a node can support and you have a fixed priority for in what order to chose, given two nodes you can automatically determine the required tunnel type and don't need to encode it in to you global description as it's a pair-wise local matter..... -:) A big part of the general problem is that all these numbers are local for each end of connections, so that you can get very asymetrical setups. 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. In the end, it boils down to a couple of things for tweaking the costs. 1. Preference of tunnel or bridge. If you prefer using the bridge, then just set the ethernet circuit cost lower than the tunnel. The tunnel can be either a multinet tunnel, or a tunnel on the cisco. 2. If the rest of your network isn't on the same ethernet as the bridge is, then you can use tweaks and control to your hearts content. 3. If the rest of your network don't have some other path out, then it don't matter if the ethernet circuit cost is high, so still feel free to tweak it as needed. I think that you should be able to tweak the costs freely almost all the time, without any issues. However, you will sometimes not get the results you are looking for if the other end, and all intermediate nodes have also set their costs based on similar views as yours. Johnny

12 years, 5 months

HECnet performance....

by roll＠Stupi.SE

On 1/9/2013 1:59 PM, Peter Lothberg wrote: On 1/8/2013 9:01 PM, Peter Lothberg wrote: Set up a (cisco) tunnel to 130.238.19.60 (and tell me your side IP address). I think it's time to come up with something a little better to track all these. Thoughts? Build a tool that takes requests and generates configs (full mesh) and pushes them to the routers. (maybe we can have the router configs in a wiki?) Great mind think alike. :) I had though pretty much exactly that. If we are going to do that, however, we should decide on a range of tunnels to use. For example on my router, 0, 1 and 2 are DECnet tunnels, 3 is my IPv6 tunnel to HE and 4 is a DECnet tunnel. Re-pushing the configs in this case would be ugly, so I think it would be wise to say, tunnels X through Y are dedicated to auto-DECnet tunnels so I can write cisco configs to deal with that appropriately. I will write something. :) -brian If you knew the capability (what kind of tunnels) a node can support and you have a fixed priority for in what order to chose, given two nodes you can automatically determine the required tunnel type and don't need to encode it in to you global description as it's a pair-wise local matter..... -:) -P

12 years, 5 months

HECnet performance....

by bqt＠softjar.se

On 2013-01-09 14:01, Steve Davidson wrote: -----Original Message----- From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On Behalf Of Peter Lothberg Sent: Wednesday, January 09, 2013 08:49 To: hecnet at Update.UU.SE Cc: hecnet at Update.UU.SE Subject: Re: [HECnet] HECnet performance.... Peter, Everything is now set up on my end. I now have 3 Cisco tunnels. The = metric is 10 on all of them. Change the metrics to match my side: Tunnel2 Cost 10 tunnel to 192.108.200.213 Cost 20 tunnel to 199.0.131.2 Cost 20 The majority of areas now route through you according to my router: Given that we have no way of setting metrics on the "bridged ethernet" links, this is the best we can do that works in the general case. I'm now trying to get the Multinet links to 192.108.200.211 59.58 (STUPI) cleaned up and re_established, right now only the link to Legato is working properly. SG1 had cost 1 on their side instead of 10. I should add that if SG1:: and LEGATO:: need to pass information, the circuit costs favor the Multinet Tunnel links between the two major US hubs. I would rather the Multinet Tunnel be used instead of using the bridge link to Sweden and then back again. I should probably just point out that there is no reason to set up the bridges to all be just connected to Update, and have everything travel through that one hub. The bridge works equally well with multiple hops. Although it does not work with a mesh, since it don't do any kind of spanning tree. Johnny

12 years, 5 months

HECnet performance....

by roll＠Stupi.SE

I was able to set the circuit cost to 10 remotely so you should be all set. Yes! **THANKS** -P

12 years, 5 months

HECnet performance....

by wonko＠4amlunch.net

On 1/9/2013 2:00 PM, Peter Lothberg wrote: On 01/08/2013 05:38 PM, Brian Hechinger wrote: Set up a (cisco) tunnel to 130.238.19.60 (and tell me your side IP address). I think it's time to come up with something a little better to track all these. Thoughts? A subdomain under some related domain (perhaps <sub>.hecnet.update.uu.se?) with the tunnel endpoints as A records? -Dave We are not limited to route DECnet packets, we could turn on IP on the tunnels, and CLNP/Phase5.. If it comes down to needing to use SNMP I'd much prefer this than exposing SNMP over the internet. -brian

12 years, 5 months

HECnet performance....

by wonko＠4amlunch.net

On 1/8/2013 5:44 PM, Dave McGuire wrote: On 01/08/2013 05:38 PM, Brian Hechinger wrote: Set up a (cisco) tunnel to 130.238.19.60 (and tell me your side IP address). I think it's time to come up with something a little better to track all these. Thoughts? A subdomain under some related domain (perhaps <sub>.hecnet.update.uu.se?) with the tunnel endpoints as A records? The problem here is that it isn't searchable. You need to know the A record name. It would be more useful for me to have it in a db. -brian

12 years, 5 months

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

by wonko＠4amlunch.net

On 1/8/2013 1:02 AM, Dave McGuire 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> So, can I get a non-priv account? something standard that we can get others to add would be nice. -brian

12 years, 5 months

subnet info, was Re: [HECnet] _PROVISIONAL_ map of HECnet, courtesy largely of Brian H.

by wonko＠4amlunch.net

On 1/9/2013 10:08 AM, Cory Smelosky wrote: On 9 Jan 2013, at 10:06, Peter Lothberg <roll at Stupi.SE> wrote: Perhaps a centralized database that maintains per-NODE info, not per-AREA info. Then that database could have a field that denotes the point of administrative control that is responsible for each node. Centralising the NODE info could solve a lot of problems and make data minin= g easier. ;) ...a wiki..... ...to be perfectly honest, a HECnet wiki isn't too bad of an idea. If it has a way to pull data from a DB, even better. :) -brian

12 years, 5 months

_PROVISIONAL_ map of HECnet, courtesy largely of Brian H.

by wonko＠4amlunch.net

On 1/9/2013 3:58 PM, Peter Lothberg wrote: Yeah more and more of us are using Ciscos to do this. We really need to find a way around this issue that doesn't involve manual maintenance of routing info. You can get the decnet routing table out of the cisco with SNMP. Need to agree on a comunity and a RO pw... Which requires "out of band" IP access. While I'm not opposed to doing that, it would be really nifty if we could do this entirely within DECnet. :) -brian

12 years, 5 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

HECnet