But you can just install another copy of multinet in a different
directory for each node and configure them separately.
Nonsense. Files in MULTINET_SPECIFIC_ROOT take
precedence over files in MULTINET_COMMON_ROOT.
I used to keep common clean, rename specific file to
common, configure it and renamed it back to specific.
--
Regards, Rok
Hi,
I looked into this a few years back and never found a solution. But you can just install another copy of multinet in a different directory for each node and configure them separately.
Dan
On 27 December 2012 21:58, <sampsa at mac.com> wrote:
I have a cluster with 3 nodes that have multinet on them.
Due to the stupidity of my router's forwarding rules, I need telnet to run on a different port on each of them, so eg:
NODE1 port 2301
NODE2 port 2302
NODE3 port 2303
Is this possible?
How do I configure this?
sampsa
On 28 Dec 2012, at 16:42, Cory Smelosky <b4 at gewt.net> wrote:
On 28 Dec 2012, at 16:35, Cory Smelosky <b4 at gewt.net> wrote:
On 28 Dec 2012, at 16:33, "Bob Armstrong" <bob at jfcl.com> wrote:
Johnny wrote:
VMS don't know about bridging, ...
Are we talking about DECnet here? If so, then that's not true. VMS is
perfectly happy with two NI interfaces. Both have the same DECnet address
and therefore the same MAC and are presumably connected to different
physical networks. It'd be a mess if you connected them both to the same
LAN segment, of course. IF the VMS node is a router, it'll happily route
DECnet traffic between the two interfaces and their associated networks.
I'm doing it now - LEGATO is configured this way.
I'm starting to think the problem lies with simh interfacing with FreeBSD's netstack/some odd OpenVPN stuff.
I fixed it. Turns out the problem resulted from accidentally becoming the router for someone else's network.
And now it's broken again.
(Feel free to ask how the hell I managed that)
Bob
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.
On 28 Dec 2012, at 16:45, Johnny Billquist <bqt at softjar.se> wrote:
On 2012-12-28 22:42, Cory Smelosky wrote:
On 28 Dec 2012, at 16:35, Cory Smelosky <b4 at gewt.net> wrote:
On 28 Dec 2012, at 16:33, "Bob Armstrong" <bob at jfcl.com> wrote:
Johnny wrote:
VMS don't know about bridging, ...
Are we talking about DECnet here? If so, then that's not true. VMS is
perfectly happy with two NI interfaces. Both have the same DECnet address
and therefore the same MAC and are presumably connected to different
physical networks. It'd be a mess if you connected them both to the same
LAN segment, of course. IF the VMS node is a router, it'll happily route
DECnet traffic between the two interfaces and their associated networks.
I'm doing it now - LEGATO is configured this way.
I'm starting to think the problem lies with simh interfacing with FreeBSD's netstack/some odd OpenVPN stuff.
I fixed it. Turns out the problem resulted from accidentally becoming the router for someone else's network.
(Feel free to ask how the hell I managed that)
Please tell. :-)
I was using OpenVPN's layer 2 bridging at my house. I have my VPS and a friend's windows server connected to it. I set up a bridge on his windows server and my DHCP server was advertising to clients on his network. I had simh instances running on that server that may've been using the default simh MAC which would explain the conflicts.
When I break networks I do so spectacularly.
Johnny
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.
On 2012-12-28 22:44, Johnny Billquist wrote:
On 2012-12-28 22:33, Bob Armstrong wrote:
Johnny wrote:
VMS don't know about bridging, ...
Are we talking about DECnet here? If so, then that's not true.
VMS is
perfectly happy with two NI interfaces. Both have the same DECnet
address
and therefore the same MAC and are presumably connected to different
physical networks. It'd be a mess if you connected them both to the same
LAN segment, of course. IF the VMS node is a router, it'll happily route
DECnet traffic between the two interfaces and their associated networks.
I'm doing it now - LEGATO is configured this way.
And I'd like to point out that what you describe is not briding... :-)
Yes, VMS can route packets. Yes, all ethernet interfaces under DECnet
phase IV will have the same MAC address. No, VMS have no idea of the
concept of a bridge. And if you bridge two ethernet segments together,
and VMS sits on both those segments, bad juju happens.
To clarify, before someone complains: all ethernet interfaces *on the same machine* will have the same MAC address.
Johnny
On 2012-12-28 22:42, Cory Smelosky wrote:
On 28 Dec 2012, at 16:35, Cory Smelosky <b4 at gewt.net> wrote:
On 28 Dec 2012, at 16:33, "Bob Armstrong" <bob at jfcl.com> wrote:
Johnny wrote:
VMS don't know about bridging, ...
Are we talking about DECnet here? If so, then that's not true. VMS is
perfectly happy with two NI interfaces. Both have the same DECnet address
and therefore the same MAC and are presumably connected to different
physical networks. It'd be a mess if you connected them both to the same
LAN segment, of course. IF the VMS node is a router, it'll happily route
DECnet traffic between the two interfaces and their associated networks.
I'm doing it now - LEGATO is configured this way.
I'm starting to think the problem lies with simh interfacing with FreeBSD's netstack/some odd OpenVPN stuff.
I fixed it. Turns out the problem resulted from accidentally becoming the router for someone else's network.
(Feel free to ask how the hell I managed that)
Please tell. :-)
Johnny
On 2012-12-28 22:33, Bob Armstrong wrote:
Johnny wrote:
VMS don't know about bridging, ...
Are we talking about DECnet here? If so, then that's not true. VMS is
perfectly happy with two NI interfaces. Both have the same DECnet address
and therefore the same MAC and are presumably connected to different
physical networks. It'd be a mess if you connected them both to the same
LAN segment, of course. IF the VMS node is a router, it'll happily route
DECnet traffic between the two interfaces and their associated networks.
I'm doing it now - LEGATO is configured this way.
And I'd like to point out that what you describe is not briding... :-)
Yes, VMS can route packets. Yes, all ethernet interfaces under DECnet phase IV will have the same MAC address. No, VMS have no idea of the concept of a bridge. And if you bridge two ethernet segments together, and VMS sits on both those segments, bad juju happens.
Johnny
Johnny Billquist <bqt at softjar.se> writes:
On 2012-12-28 21:00, Cory Smelosky wrote:
Hello!
%NCP-W-OPEFAI, Operation failure
%SYSTEM-F-IVADDR, invalid media address
%NCP-W-UNRCMP, Unrecognized component , Circuit
Circuit = QNA-1
However:
XQ, address=20001920-2000192F*, vector=250, MAC=AA:00:04:00:15:24, type=DELQA-T, poll=100, attached to em0
===================================================^^^^^^^^^^^^^^^^^
XQB, address=20001930-2000193F*, vector=2E0, MAC=AA:00:04:00:15:24, type=DELQA-T, poll=100, attached to tap0
====================================================^^^^^^^^^^^^^^^^^
DECnet modifies the ethernet address of the interface. If these sit
on a common ethernet, you're going to have problems! Perhaps, that's
what you are seeing. Trying setting the second circuit to a unique
address. Try address 9.22 on the second ethernet circuit.
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
Well I speak to machines with the voice of humanity.
On 28 Dec 2012, at 16:35, Cory Smelosky <b4 at gewt.net> wrote:
On 28 Dec 2012, at 16:33, "Bob Armstrong" <bob at jfcl.com> wrote:
Johnny wrote:
VMS don't know about bridging, ...
Are we talking about DECnet here? If so, then that's not true. VMS is
perfectly happy with two NI interfaces. Both have the same DECnet address
and therefore the same MAC and are presumably connected to different
physical networks. It'd be a mess if you connected them both to the same
LAN segment, of course. IF the VMS node is a router, it'll happily route
DECnet traffic between the two interfaces and their associated networks.
I'm doing it now - LEGATO is configured this way.
I'm starting to think the problem lies with simh interfacing with FreeBSD's netstack/some odd OpenVPN stuff.
I fixed it. Turns out the problem resulted from accidentally becoming the router for someone else's network.
(Feel free to ask how the hell I managed that)
Bob
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.
On 28 Dec 2012, at 16:33, "Bob Armstrong" <bob at jfcl.com> wrote:
Johnny wrote:
VMS don't know about bridging, ...
Are we talking about DECnet here? If so, then that's not true. VMS is
perfectly happy with two NI interfaces. Both have the same DECnet address
and therefore the same MAC and are presumably connected to different
physical networks. It'd be a mess if you connected them both to the same
LAN segment, of course. IF the VMS node is a router, it'll happily route
DECnet traffic between the two interfaces and their associated networks.
I'm doing it now - LEGATO is configured this way.
I'm starting to think the problem lies with simh interfacing with FreeBSD's netstack/some odd OpenVPN stuff.
Bob
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.
