Mark,
This is how SGC:: works today. It is running NetBSD, SimH, and VAX/VMS 7.3. Save yourself tons of grief and use two NICs. One NIC is for NetBSD (IP) the other is for VMS (DECnet and/or IP). Bring the second NIC up under NetBSD but do NOT assign an IP address to it. If you wish you can run the bridge on this same machine or run Multinet with its tunnels.
I can give you access to the SGC:: host machine so that you take a look if you wish. Just let me know.
Regards,
-Steve
-----Original Message-----
From: owner-hecnet at Update.UU.SE on behalf of Mark Benson
Sent: Thu 6/30/2011 04:33
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] SIMH VAX Project
Has anyone ever run SIMH out of NetBSD? I'm curious as to how to link the Ethernet to the LAN device (ra0) on my D410 board.
It fails if I use:
attach xq ra0
OR
attach xq /dev/ra0
So not sure what to assign it to...
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
On 30 Jun 2011, at 09:05, Fausto Saporito <fausap at libero.it> wrote:
Hello,
the TCPIP package is on the CD, and you have to install after the installation.
Before, you have to change some system parameters.
A good guide is http://www.wherry.com/gadgets/retrocomputing/vax-simh.html
Here you can find all the instructions to set up your sistem.
Fausto
On 30.6.2011 17:51, Mark Benson wrote:
On 30 Jun 2011, at 14:40, Johnny Billquist<bqt at softjar.se> wrote:
. However, most people just find it overly complex to manage compared to DECnet IV, without any tangible benefits.
I will need to look at reconfiguring my Alpha I think. Contemplating changing the disks around in it anyway so might try and get it right on re-install.
You do as you please, of course. I think Johnny is somewhat exaggerating about the DECnet-plus management. It is not _that_ difficult to handle. Many others claim that it is too complex, but IMHO the point is that the design and philosophy is different from DECnet Phase IV.
It's like having a box camera and a system camera; you can take pictures with both, but when you want to fine tune your pictures or take pictures in other conditions than in sunshine you need a system camera.
You have a much better and accurate control of what's happening within your DECnet stack in every layer. In the integrated implementation you have OSI as well right out-of-the-box. I've never seen another OSI implementation which has such a detailed and comprehesive management. It is a masterpiece. No wonder it was used in really big companies. DEC used it in its worldwide internal network (called the Easynet), of course. And it worked like a charm.
In fact the Phase IV was way too limited for bigger networks. In many companies there were needed to do tricks to get all needed nodes attached to the network. After Phase V was implemented, all limitations were gone.
Then came TCP/IP and mixed up everything, but that's another story. :)
Why not play around with DECnet-plus when you already have it installed and working?
As for the "problems" with NetBSD, I think I saw that you figured out that simh needs to be run as root for it to work. That should be pretty obvious, since it would be a big security problem if everyone was allowed to sniff the ethernet cables at will. Also, it is "ra0" and not "/dev/ra0". Ethernet device names are not paths that exist in the file system. It should be obvious if you actually do an ls on /dev... There isn't any /dev/ra0 (well, unless you happen to run a VAX with NetBSD, in which case ther
e would be a /dev/ra0[a-h] for a disk drive.
NetBSD's /dev is very confusing - a whole lot of nodes exist for hardware that isn't there (as far as I can see).
I was clutching at straws. :)
Yes, both simh and the machine itself can share the same ethernet port, with the possible problem of talking between those two entities. But from other machines, there will not be any problems (this is not an aspect of the pcap library, but actually a thing related to the Berkley packet filter (bpf) which pcap use, and also to the underlying ethernet driver and possibly the hardware itself).
However, there is in general a bigger problem with simh, NetBSD and DECnet, in that DECnet expects to be able to change the MAC address of the ethernet interface
SIMH uses a separate MAC address that is specified in the bootstrap configuration. I'd imagine (I don't know for sure) that DECNet's MAC control would alter this value, not require altering the hardware embedded MAC of the NIC. The two instances appear to outside Ethernet devices as separate entities as far as I can tell.
As for compiling, yes, the pcap library have to be around already at compile time, or the compilation will fail (obviously). Not withstanding shared libraries and all that, the header files, which are also installed along with the pcap library itself, are needed for compilation. And at link time, the library is checked for some information at that time also. That said, libpcap is a standard component of NetBSD, and there is no need to install another version from pkgsrc.
It may have pulled it in as a dependancy via pkgsrc. I know it compiled and installed 4 packages.
Well, not really. Linux and Unix are written by different people, but they are based on the same basic design. VMS and WNT was designed by the same person, but not designed in the same way.
Anyone using one Unix system will feel pretty much at home with any other Unix system. Yes, some details differs, but almost everything will be the same, down to using the same shell, similary file system layout, and so on... Try that between VMS and WNT if you dare... :-)
It was only a joke. ;) ;)
I hostnamed my new Windows PC 'cutler' - I needed a name and as it's running a spiritual successor of NT I thought it was appropriate ;)
On 06/30/11 16:51, Mark Benson wrote:
On 30 Jun 2011, at 14:40, Johnny Billquist<bqt at softjar.se> wrote:
. However, most people just find it overly complex to manage compared to DECnet IV, without any tangible benefits.
I will need to look at reconfiguring my Alpha I think. Contemplating changing the disks around in it anyway so might try and get it right on re-install.
Changing from DECnet-PLUS to DECnet IV is actually simple. You just uninstall one product, and install the other one.
(And then you need to configure it, of course...)
As for the "problems" with NetBSD, I think I saw that you figured out that simh needs to be run as root for it to work. That should be pretty obvious, since it would be a big security problem if everyone was allowed to sniff the ethernet cables at will. Also, it is "ra0" and not "/dev/ra0". Ethernet device names are not paths that exist in the file system. It should be obvious if you actually do an ls on /dev... There isn't any /dev/ra0 (well, unless you happen to run a VAX with NetBSD, in which case there would be a /dev/ra0[a-h] for a disk drive.
NetBSD's /dev is very confusing - a whole lot of nodes exist for hardware that isn't there (as far as I can see).
It's the traditional Unix way. :-)
Linux have moved into having a hierarchy, and sometimes also creating and deleting entries dynamically. As do some other Unix variants.
In NetBSD, they are created for the "just in case" scenario. And it's very flat. Actually much easier to understand and manage, if you ask me.
Any entries you don't need, you can just delete them.
There is a MAKEDEV script in /dev which can recreate them for you, if you ever need to. Or you can do it by hand, of course...
I was clutching at straws. :)
Nothing wrong with that. :-)
Yes, both simh and the machine itself can share the same ethernet port, with the possible problem of talking between those two entities. But from other machines, there will not be any problems (this is not an aspect of the pcap library, but actually a thing related to the Berkley packet filter (bpf) which pcap use, and also to the underlying ethernet driver and possibly the hardware itself).
However, there is in general a bigger problem with simh, NetBSD and DECnet, in that DECnet expects to be able to change the MAC address of the ethernet interface
SIMH uses a separate MAC address that is specified in the bootstrap configuration. I'd imagine (I don't know for sure) that DECNet's MAC control would alter this value, not require altering the hardware embedded MAC of the NIC. The two instances appear to outside Ethernet devices as separate entities as far as I can tell.
Hmm. I haven't checked, but I guess it just puts the ethernet interface in promiscuous mode, and does all the packet matching itself then.
That should work just fine, for the cost of some extra cpu cycles.
As for compiling, yes, the pcap library have to be around already at compile time, or the compilation will fail (obviously). Not withstanding shared libraries and all that, the header files, which are also installed along with the pcap library itself, are needed for compilation. And at link time, the library is checked for some information at that time also. That said, libpcap is a standard component of NetBSD, and there is no need to install another version from pkgsrc.
It may have pulled it in as a dependancy via pkgsrc. I know it compiled and installed 4 packages.
If you build simh from pkgsrc, then yes. simh have a depencency on libpcap >= 0.6
I'm somewhat intruiged that they actually made it dependant on libpcap from pkgsrc instead of using the system libpcap. But that is another problem, for another day... :-)
Well, not really. Linux and Unix are written by different people, but they are based on the same basic design. VMS and WNT was designed by the same person, but not designed in the same way.
Anyone using one Unix system will feel pretty much at home with any other Unix system. Yes, some details differs, but almost everything will be the same, down to using the same shell, similary file system layout, and so on... Try that between VMS and WNT if you dare... :-)
It was only a joke. ;) ;)
Actually, I'm using yet another OS written by Dave Cutler every day, and that is RSX... :-) (And it wasn't just designed by him, his name is all over the sources...)
I hostnamed my new Windows PC 'cutler' - I needed a name and as it's running a spiritual successor of NT I thought it was appropriate ;)
:-)
Johnny
On 30 Jun 2011, at 14:40, Johnny Billquist <bqt at softjar.se> wrote:
. However, most people just find it overly complex to manage compared to DECnet IV, without any tangible benefits.
I will need to look at reconfiguring my Alpha I think. Contemplating changing the disks around in it anyway so might try and get it right on re-install.
As for the "problems" with NetBSD, I think I saw that you figured out that simh needs to be run as root for it to work. That should be pretty obvious, since it would be a big security problem if everyone was allowed to sniff the ethernet cables at will. Also, it is "ra0" and not "/dev/ra0". Ethernet device names are not paths that exist in the file system. It should be obvious if you actually do an ls on /dev... There isn't any /dev/ra0 (well, unless you happen to run a VAX with NetBSD, in which case there would be a /dev/ra0[a-h] for a disk drive.
NetBSD's /dev is very confusing - a whole lot of nodes exist for hardware that isn't there (as far as I can see).
I was clutching at straws. :)
Yes, both simh and the machine itself can share the same ethernet port, with the possible problem of talking between those two entities. But from other machines, there will not be any problems (this is not an aspect of the pcap library, but actually a thing related to the Berkley packet filter (bpf) which pcap use, and also to the underlying ethernet driver and possibly the hardware itself).
However, there is in general a bigger problem with simh, NetBSD and DECnet, in that DECnet expects to be able to change the MAC address of the ethernet interface
SIMH uses a separate MAC address that is specified in the bootstrap configuration. I'd imagine (I don't know for sure) that DECNet's MAC control would alter this value, not require altering the hardware embedded MAC of the NIC. The two instances appear to outside Ethernet devices as separate entities as far as I can tell.
As for compiling, yes, the pcap library have to be around already at compile time, or the compilation will fail (obviously). Not withstanding shared libraries and all that, the header files, which are also installed along with the pcap library itself, are needed for compilation. And at link time, the library is checked for some information at that time also. That said, libpcap is a standard component of NetBSD, and there is no need to install another version from pkgsrc.
It may have pulled it in as a dependancy via pkgsrc. I know it compiled and installed 4 packages.
Well, not really. Linux and Unix are written by different people, but they are based on the same basic design. VMS and WNT was designed by the same person, but not designed in the same way.
Anyone using one Unix system will feel pretty much at home with any other Unix system. Yes, some details differs, but almost everything will be the same, down to using the same shell, similary file system layout, and so on... Try that between VMS and WNT if you dare... :-)
It was only a joke. ;) ;)
I hostnamed my new Windows PC 'cutler' - I needed a name and as it's running a spiritual successor of NT I thought it was appropriate ;)
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
Hi. Apologies for coming in late. I'll try to answer several mails and questions at the same time here...
On 06/30/11 09:10, Mark Benson wrote:
On a semi-related note I noticed something installing 7.3 that it said not to install DECNet-Plus if you want to use DECNet-PhaseIV. Since HECNet is PhaseIV I didn't install Plus on the VAX EMulator, but I *did* IIRC on the Alpha. Will the Alpha need a re-install or reconfiguration of some kind to run on HECNet?
DECnet-Plus is pretty much DECnet IV compatible as far as interoperability goes. What differs is the way you manage the whole thing, as well as all the extra stuff that is in DECnet-Plus. But you can run DECnet-Plus just fine on HECnet, and I think one or two actually do. However, most people just find it overly complex to manage compared to DECnet IV, without any tangible benefits.
As for the "problems" with NetBSD, I think I saw that you figured out that simh needs to be run as root for it to work. That should be pretty obvious, since it would be a big security problem if everyone was allowed to sniff the ethernet cables at will. Also, it is "ra0" and not "/dev/ra0". Ethernet device names are not paths that exist in the file system. It should be obvious if you actually do an ls on /dev... There isn't any /dev/ra0 (well, unless you happen to run a VAX with NetBSD, in which case there would be a /dev/ra0[a-h] for a disk drive.
Yes, both simh and the machine itself can share the same ethernet port, with the possible problem of talking between those two entities. But from other machines, there will not be any problems (this is not an aspect of the pcap library, but actually a thing related to the Berkley packet filter (bpf) which pcap use, and also to the underlying ethernet driver and possibly the hardware itself).
However, there is in general a bigger problem with simh, NetBSD and DECnet, in that DECnet expects to be able to change the MAC address of the ethernet interface, and the ethernet interface on NetBSD does not have a way of doing this. I'm not sure if simh have a workaround for that, but I would be suspicious that maybe DECnet will not work under simh on NetBSD. IP will work fine, however, since it does not try to do such a trick.
As for compiling, yes, the pcap library have to be around already at compile time, or the compilation will fail (obviously). Not withstanding shared libraries and all that, the header files, which are also installed along with the pcap library itself, are needed for compilation. And at link time, the library is checked for some information at that time also. That said, libpcap is a standard component of NetBSD, and there is no need to install another version from pkgsrc.
> Hi Mark, you do know that one of the symptoms of being a VMS bigot is that unixes look alike :-)
That's kinda like confusing VMS and Windows NT because they both have something to do with Dave Cutler and associates ;)
Well, not really. Linux and Unix are written by different people, but they are based on the same basic design. VMS and WNT was designed by the same person, but not designed in the same way.
Anyone using one Unix system will feel pretty much at home with any other Unix system. Yes, some details differs, but almost everything will be the same, down to using the same shell, similary file system layout, and so on... Try that between VMS and WNT if you dare... :-)
Johnny
Security on BSD distributions is tight, you might need to reduce the security policies on the system. Libpcap turns the ethernet device in promiscuous mode and that may be considered harmful.
Figures. You can only run SIMH with networking as root. Might explain it.
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
On 30 Jun 2011, at 10:56, hvlems at zonnet.nl wrote:
Hans
Verzonden vanaf mijn draadloze BlackBerry -toestel
-----Original Message-----
From: Mark Benson <md.benson at gmail.com>
Sender: owner-hecnet at Update.UU.SE
Date: Thu, 30 Jun 2011 10:26:19
To: hecnet at Update.UU.SE<hecnet at Update.UU.SE>
Reply-To: hecnet at Update.UU.SESubject: Re: [HECnet] SIMH VAX Project
"File not found" I think. Like it couldn't pick up the device.
Re: Sharing a port, I don't need to talk to the host OS from the VAX emulation, I just need to access both from a third host.
I will have to compile libpcap from pkgsrc but that shouldn't be an issue unless pcap needs to be present at compile time. If that is the case I just need to recompile SIMH so again no big deal.
FWIW, NetBSD is not a flavour of Linux either ;) ;)
<this is intended as a lighthearted statement>
It's all good fun!
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
On 30 Jun 2011, at 09:53, hvlems at zonnet.nl wrote:
What error message did you get?
Mark. if you want me to hook you up with some HPUX media I'd be happy to.
I'd be very happy if you could - I need a suitable set of install and applications media for IA64 - preferably 11i v2 or later.
I have a zx6000 like yours but with no skins or stand. Nice machine but boy does it throw out some heat!
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
On 30/06/11 11:58, Mark Benson wrote:
Hi Mark, you do know that one of the symptoms of being a VMS bigot is that unixes look alike :-)
That's kinda like confusing VMS and Windows NT because they both have something to do with Dave Cutler and associates ;)
Actually, when I typed the reply I had forgotten which host os it was. NetBSD is pretty easy to install IIRC, had it installed on a VAX and an Alpha. Never did anything with it.
I have run it on a multitude of architectures. In all cases it's insanely easy to bootstrap. Also they start you out with just the basic OS, TCP/IP access, SSH and no crap to get in the way. Thus it was a natural choice for me as a minimal framework to run SIMH on.
These days all vaxes run VMS as do most alphas. Three alphas run Tru64 though.
I have a policy if not running generic OSs on specialist hardware. If I want a generic OS I use a generic computer I.e. x86_64).
All my non-x86 machines run their intended OSs apart for my zx6000 - I don't need a third VMS machine and I don't have the HP-UX 11i v2 IA64 media that goes with it :(
File not found is a curious error message. It's as if there is a typo in a .conf file in /etc?
Unlikely. The Ethernet works fine :)
/dev/ra0 is a strange name (for me and I know I lead a sheltered life ;-), what kind of hardware is it (1000 Mb/s ethernet perhaps).
Whatever is on the D410PT motherboard. Probably realtek (hence the ra0 - follows old UNIX convention of using vendor names for devices ala wd0 being hard drive storage named for Western Digital)
Or possibly because /dev/ra0 doesn't exist or cannot be accessed by libpcap (does it need root privs to run)?
I strongly suspect libpcap is missing. Pretty much everything is optional in NetBSD!
Mark. if you want me to hook you up with some HPUX media I'd be happy to.
Mark.
Hi Mark, you do know that one of the symptoms of being a VMS bigot is that unixes look alike :-)
That's kinda like confusing VMS and Windows NT because they both have something to do with Dave Cutler and associates ;)
Actually, when I typed the reply I had forgotten which host os it was. NetBSD is pretty easy to install IIRC, had it installed on a VAX and an Alpha. Never did anything with it.
I have run it on a multitude of architectures. In all cases it's insanely easy to bootstrap. Also they start you out with just the basic OS, TCP/IP access, SSH and no crap to get in the way. Thus it was a natural choice for me as a minimal framework to run SIMH on.
These days all vaxes run VMS as do most alphas. Three alphas run Tru64 though.
I have a policy if not running generic OSs on specialist hardware. If I want a generic OS I use a generic computer I.e. x86_64).
All my non-x86 machines run their intended OSs apart for my zx6000 - I don't need a third VMS machine and I don't have the HP-UX 11i v2 IA64 media that goes with it :(
File not found is a curious error message. It's as if there is a typo in a .conf file in /etc?
Unlikely. The Ethernet works fine :)
/dev/ra0 is a strange name (for me and I know I lead a sheltered life ;-), what kind of hardware is it (1000 Mb/s ethernet perhaps).
Whatever is on the D410PT motherboard. Probably realtek (hence the ra0 - follows old UNIX convention of using vendor names for devices ala wd0 being hard drive storage named for Western Digital)
Or possibly because /dev/ra0 doesn't exist or cannot be accessed by libpcap (does it need root privs to run)?
I strongly suspect libpcap is missing. Pretty much everything is optional in NetBSD!
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
Hi Mark, you do know that one of the symptoms of being a VMS bigot is that unixes look alike :-)
Actually, when I typed the reply I had forgotten which host os it was. NetBSD is pretty easy to install IIRC, had it installed on a VAX and an Alpha. Never did anything with it. These days all vaxes run VMS as do most alphas. Three alphas run Tru64 though.
File not found is a curious error message. It's as if there is a typo in a .conf file in /etc?
/dev/ra0 is a strange name (for me and I know I lead a sheltered life ;-), what kind of hardware is it (1000 Mb/s ethernet perhaps).
Or possibly because /dev/ra0 doesn't exist or cannot be accessed by libpcap (does it need root privs to run)?
Security on BSD distributions is tight, you might need to reduce the security policies on the system. Libpcap turns the ethernet device in promiscuous mode and that may be considered harmful.
Hans
Verzonden vanaf mijn draadloze BlackBerry -toestel
-----Original Message-----
From: Mark Benson <md.benson at gmail.com>
Sender: owner-hecnet at Update.UU.SE
Date: Thu, 30 Jun 2011 10:26:19
To: hecnet at Update.UU.SE<hecnet at Update.UU.SE>
Reply-To: hecnet at Update.UU.SESubject: Re: [HECnet] SIMH VAX Project
"File not found" I think. Like it couldn't pick up the device.
Re: Sharing a port, I don't need to talk to the host OS from the VAX emulation, I just need to access both from a third host.
I will have to compile libpcap from pkgsrc but that shouldn't be an issue unless pcap needs to be present at compile time. If that is the case I just need to recompile SIMH so again no big deal.
FWIW, NetBSD is not a flavour of Linux either ;) ;)
<this is intended as a lighthearted statement>
It's all good fun!
--
Mark Benson
http://markbenson.org/bloghttp://twitter.com/MDBenson
On 30 Jun 2011, at 09:53, hvlems at zonnet.nl wrote:
What error message did you get?
