[HECnet] *** Very important information ***

[Johnny Billquist]

Hi, all. This mail contains some very important information that affects 
a fair number of people and services on HECnet. So please read through 
it, and if you need to take action, please do this in a timely manner. 
I'm here to help with whatever, but I can only work on things from my 
end of things.


Background:

Update is a computer club at Uppsala University. As such, it has enjoyed 
a lot of support from the University over a long time. However, recent 
development have now led us to a point where the University is no longer 
willing to support Update.
The main way the support have been provided is through a computer room 
where machines have been running, electricity, and network connectivity.

All of this is now going away. Basically, and the end of this year, 
Update will have to have moved out of their current location.

This means that any machine Update currently have running will need to 
move. Update have found a location of their own that they now started to 
rent, and most things will be moving in a few weeks from now. There are 
also unfortunately a need to reduce the size of Update's collection of 
hardware since the size of the collection is just too big for Update to 
afford to keep. You might hear more on that separately, but I'm not 
going to go into that any further than saying that Magica and other 
large iron is currently safe, and will be moved. But it's going to be 
difficult to run these machines for the time being.


Primary effects:

Mim.Update.UU.SE, Psilo.Update.UU.SE and any other resources used by 
HECnet will go offline. I'm not sure about the exact date, but expect 
this to happen in a couple of weeks from now.

This means any Multinet links to MIM will go down, and will not come 
back up. Any bridge connections to Psilo will also go away.
Also, the HECnet mailing list will stop working.

Update is working on getting services back up and running in their new 
location, but exactly in which form we are able to is unclear, and it is 
also unclear what network capacity there will be, or when anything will 
be back.


Changes happening on HECnet:

Because of the things mentioned above, there is a plan in place to keep 
HECnet up and running. Thanks to Peter Lothberg, a lot of functions and 
services currently at Update will move to Stupi.

MIM:: (Mim.Update.UU.SE) has already moved to Stupi, and is now known as 
Mim.Stupi.NET. Still the same nodename on HECnet.
The old Mim (Mim.Update.UU.SE) is still up for the time being, as XMIM::

Any Multinet links to Mim.Update.UU.SE are going to have to move. There 
will be two alternatives:

. HECnet-1-1023.Stupi.NET is a PyDECnet router which I intend to be a 
major hub for HECnet. Located in Stockholm with good network capacity.

. Mim.Stupi.NET, which is just the same machine, and can accept the same 
links.


Any bridge connection to Psilo.Update.UU.SE are going to have to move. 
There is one alternative:

. HECnet-1-1023.Stupi.NET. Same router that takes Multinet links. As 
mentioned above, I aim for this to be a major hub going forward.



XMIM:: will be around for a few weeks, but will go down in the not 
distant future. All current Multinet bridges to it will have to migrate 
in order to keep working. Else those links will soon stop.


MIM:: will continue service HECnet with the nodename database, just like 
before. It also will continue to act as a mail gateway for anyone using 
that feature. Mail from inside HECnet to the Internet will continue to 
be handled as MIM::<user at internet.host>
Mail from Internet to HECnet will need to be addressed as 
<node>::<user>@Mim.Stupi.Net

Web services at Mim will continue as well, but obviously at 
Mim.Stupi.NET. Same with ftp, rpm, and all other things.


In order to reduce the number of issues that might pop up, I am going to 
try and get Mim.Update.UU.SE setup as a CNAME to Mim.Stupi.NET, so that 
most things will continue to work even using the old hostname, but there 
might be hiccups here. Be aware.


The mailing list I'm still looking into how to handle. I'll get back on 
that soon.


Next steps:

I want that people have have links to Update to contact me, and we can 
move the links over to new endpoints. Keep in mind that I might take a 
day or two to respond to each individual. But I'll try to move fast as I 
don't know when things will stop operating in the current setup.

Also, any issues, problems, funny stuff, or whatever, please reach out 
to me. I'm also going to keep an eye open for routing becoming odd in 
the face of things here, and will poke people when I see things that 
just look wrong.
Of course, I have no insight into any Cisco equipment, so there I will 
just have to hope and trust that each person running such gear have 
things set up in a reasonable way.



Well, that's about all I can think of right now.
Now you all know what is happening, and anyone having links to Update 
will need to take action. Anyone using other services from machines at 
Update might also see problems as this all happens.



And once more, many thanks to Peter Lothberg, who have been very helpful 
in finding a solution to this problem.

(And yes, I wish Uppsala University had done things differently, but 
that is not anything I am in any position to affect, or make much 
comments on.)

   Johnny

-- 
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: bqt at softjar.se             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol