Does anyone have sources to RMS-11 V2.0. Specifically for RSX, in case they differ.
I have a tricky problem. I've written a device driver that pretends to look like a file system without using any ACP. It implements most functions an ACP normally implement. It also have some "odd" attributes set on the device to represent both what it actually is, and also to make FCS do "the right thing".
So, yes, using FCS, I can access this device just like any other, including opening files and so on... However, RMS-11 is not so cooperative, and I guess it means that I'm either not implementing some ACP function in enough detail to make RMS-11 happy, or else some device attribute it not set right, causing RMS-11 to treat the whole thing in a way that is too restricted, compared to FCS, in which case I want to figure out exactly how to set attributes to make RMS-11 do the right thing as well.
And just to explain what I'm doing in a little more detail:
I've actually done this for two devices. The first is the DOS device presented by E11 and other emulators. It's a device that gives access to the host file system.
The second device is actually TCP/IP. I can use normal tools and FCS functions to access the internet, just as easy as any other file. But not so, using RMS... :-(
As an example:
.pip ti:=tc:"130.238.19.25";21
220- ===============================================================
220- = Welcome to the Update Computer Club FTP archives! =
220- ===============================================================
220-
220- This server is maintained by the computer club Update at
220- Uppsala University in Uppsala, Sweden.
220-
220- You can find all the good stuff in the /pub directory.
220-
220- Please be advised that the club provides this service without
220- any guarantees, and we take no responsibility for any harm that
220- may come to you or your system from the use of this server, or
220- files obtained from this server.
220-
220- We do, however, try to ensure that everything here is correct.
220- If you have problems with the server, please contact the
220- FTP administrator ftpadm at Update.UU.SE.
220-
220
PIP -- I/O error on input file
TC0:[IPLIB]130238019.025;21 -- Timeout on request
.
Now, it would be awesome if I could make this play using RMS as well, since then I could easily write servers using BASIC+2 or other languages. FORTRAN 77 as well as C do also have FCS interfaces, so I can already use those, assuming I'm happy with limiting myself to FCS. But I want it all (of course). So - anyone who might be able to help?
(And no, the TCP/IP is not available for distribution yet, but might be in the future, depending on interest and so 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
Steve,
I see your multinet tunnel to GORVAX is down - is there a problem your side or mine, you think?
Sampsa
On 6 Sep 2011, at 17:03, Steve Davidson wrote:
Mark,
If you need and (remote) help with this...
Let me know. I will make myself available as necessary.
-Steve
________________________________
From: owner-hecnet at Update.UU.SE on behalf of Mark Wickens
Sent: Tue 9/6/2011 02:55
To: hecnet at Update.UU.SE
Subject: [HECnet] DEC Legacy Event is nearly upon us...
Just a reminder that the DEC Legacy Event is nearly upon us!
October 8/9 in Windermere, more information here: http://declegacy.org.uk <http://declegacy.org.uk/>
The exhibits page gives a summary of the hardware that will be making an
appearance: http://wickensonline.co.uk/declegacy/hardware
I'm in the process of securing an internet connection. Hopefully I'll be
connecting a couple of my boxes to HECnet over the course of the
weekend, to show participants what they're missing out on.
Regards, Mark.
<winmail.dat>
Mark,
If you need and (remote) help with this...
Let me know. I will make myself available as necessary.
-Steve
________________________________
From: owner-hecnet at Update.UU.SE on behalf of Mark Wickens
Sent: Tue 9/6/2011 02:55
To: hecnet at Update.UU.SE
Subject: [HECnet] DEC Legacy Event is nearly upon us...
Just a reminder that the DEC Legacy Event is nearly upon us!
October 8/9 in Windermere, more information here: http://declegacy.org.uk <http://declegacy.org.uk/>
The exhibits page gives a summary of the hardware that will be making an
appearance: http://wickensonline.co.uk/declegacy/hardware
I'm in the process of securing an internet connection. Hopefully I'll be
connecting a couple of my boxes to HECnet over the course of the
weekend, to show participants what they're missing out on.
Regards, Mark.
Lots of Unices use "partition" names to designate part of all of a physical device; for example on NetBSD the suffix "d" by convention means "the whole disk". On the other hand, on Linux no suffix is the whole disk, and letter suffixes mean (real) partitions.
I didn't know that CDRoms could be partitioned. But you can check this easily: "cat /proc/partitions" shows you all the partitions that the OS can see (on all the visible disks).
paul
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On Behalf Of Mark Benson
Sent: Tuesday, September 06, 2011 2:22 AM
To: hecnet at Update.UU.SE
Subject: Re: [HECnet] SIMH on CentOS 5.6
On 6 Sep 2011, at 06:52, hvlems at zonnet.nl wrote:
The dd command ought to produce a bootable copy of the cdrom. On Tru64
the correct partition is If=/dev/disks/ cdrom0c
When I dd'd off a copy of the VAX VMS CD-ROM it didn't work for me initially either. Now I can't remember which way round it was now, but I think I dd'd the whole CD-ROM and it didn't work and then I dd'd off only the *partition* with the data in and that worked fine. Either that or the other way around. From the looks of the quote above you only need the partition.
Memory is terrible, sorry :)
--
Mark Benson
My Blog:
<http://markbenson.org/blog>
Follow me on Twitter:
http://twitter.com/mdbenson
"Never send a human to do a machine's job..."
On Mon, 5 Sep 2011 22:41:35 -0700, you wrote:
Does anyone have any pointers for making an ISO image of the CD? Am I doing
it wrong? I would like to make a backup of the disk because I'm already on
my second physical copy and would like to back it up before I lose it again
You may already know, but... Please note that a VMS CD-ROM bootable image is
not ISO 9660 formatted. So don't be misleaded if a correctly made image does
not mount on Linux or whatever (or does not show up on Windows): don't take
that as a test! The only way to check for a VMS CD-ROM image integrity is by
having it booted or mounted in VMS.
There is also a vmscd utility that can be used to inspect and extract data
from such an image on Linux. Just look for it with Google.
There is no difference between a VMS CD-ROM bootable image and a classical
VMS hard disk image: they are both structured as an ODS-2 filesystem.
Anyway, VMS can also mount and use ISO 9660 CD-ROM data discs and images.
HTH, :-)
G.
P.S. Itanium-bootable images are different.
Just a reminder that the DEC Legacy Event is nearly upon us!
October 8/9 in Windermere, more information here: http://declegacy.org.uk
The exhibits page gives a summary of the hardware that will be making an appearance: http://wickensonline.co.uk/declegacy/hardware
I'm in the process of securing an internet connection. Hopefully I'll be connecting a couple of my boxes to HECnet over the course of the weekend, to show participants what they're missing out on.
Regards, Mark.
On 6 Sep 2011, at 07:30, Ian McLaughlin wrote:
When I dd'd off a copy of the VAX VMS CD-ROM it didn't work for me initially either. Now I can't remember which way round it was now, but I think I dd'd the whole CD-ROM and it didn't work and then I dd'd off only the *partition* with the data in and that worked fine. Either that or the other way around. From the looks of the quote above you only need the partition.
Memory is terrible, sorry :)
On my system (CentOS 5.6) I see the /dev/cdrom device which is symbolically linked to /dev/hda (which makes sense - it's the only IDE device I have in the machine). There is no hda1 or other devices for the partitions. fdisk /dev/hda comes back with an invalid partition table.
I remember how it went now - I was doing it on a Mac which uses a BSD disk structure. I dd'd the disk device (/dev/disk6) and it wouldn't work. I then dd'd the slice that reported as the actual volume (/dev/disk6s0) and it worked a charm.
I don't know how that's represented in Linux, however. Also fdisk won't show a partition table for a CD-ROM, I don't believe, because they are not structured as hard drive partitions. I put my VMS VAX CD-ROM in my Ubuntu box and the Disk Utility reports it at /dev/sr0 which is a SCSI ID but my Microserver has a weird disk controller.
I honestly don't know how to approach it unless you have something like gnuparted to hand to look at the disk?
--
Mark Benson
My Blog:
<http://markbenson.org/blog>
Follow me on Twitter:
http://twitter.com/mdbenson
"Never send a human to do a machine's job..."
On 2011-09-05, at 11:21 PM, Mark Benson wrote:
On 6 Sep 2011, at 06:52, hvlems at zonnet.nl wrote:
The dd command ought to produce a bootable copy of the cdrom. On Tru64 the correct partition is
If=/dev/disks/ cdrom0c
When I dd'd off a copy of the VAX VMS CD-ROM it didn't work for me initially either. Now I can't remember which way round it was now, but I think I dd'd the whole CD-ROM and it didn't work and then I dd'd off only the *partition* with the data in and that worked fine. Either that or the other way around. From the looks of the quote above you only need the partition.
Memory is terrible, sorry :)
On my system (CentOS 5.6) I see the /dev/cdrom device which is symbolically linked to /dev/hda (which makes sense - it's the only IDE device I have in the machine). There is no hda1 or other devices for the partitions. fdisk /dev/hda comes back with an invalid partition table.
Ian
On 6 Sep 2011, at 06:52, hvlems at zonnet.nl wrote:
The dd command ought to produce a bootable copy of the cdrom. On Tru64 the correct partition is
If=/dev/disks/ cdrom0c
When I dd'd off a copy of the VAX VMS CD-ROM it didn't work for me initially either. Now I can't remember which way round it was now, but I think I dd'd the whole CD-ROM and it didn't work and then I dd'd off only the *partition* with the data in and that worked fine. Either that or the other way around. From the looks of the quote above you only need the partition.
Memory is terrible, sorry :)
--
Mark Benson
My Blog:
<http://markbenson.org/blog>
Follow me on Twitter:
http://twitter.com/mdbenson
"Never send a human to do a machine's job..."