peter@csd4.csd.uwm.edu (Peter J Diaz de Leon) (09/06/89)
We have at work some ATT 3B1's and a 20MHz ATT 60386 box. Both are running ATT Sysyem V. Is there any way to get a floppy that was formated on a 3B1 read on a 60386 and vise versa. Thanks Peter ============================================================================== ARPA: peter@csd4.csd.uwm.edu USMAIL: Peter J. Diaz de Leon peter@cvax.cs.uwm.edu 7411 W. Warnimont Ave. Milwaukee, WI. 53220 ICBM: 43 4 58 N / 87 55 52 W ==============================================================================
paone@aramis.rutgers.edu (Phil Paone) (09/07/89)
It is easy to read 3B1 floppies on the AT&T 6386. One of the devices for the floppy does the proper density and everything for the 3B1. I don't have the manuals in front of me, but the 3.2 sysadm guide that comes with the 386 tells the devices to use for the UNIX PC and the 3B2 disk formats. -- Phil Paone attmail!ppaone !rutgers.edu!topaz.edu!ppaone paone@topaz.rutgers.edu "Dinna ya know a jailbreak when ya see it?"
psfales@cbnewsc.ATT.COM (Peter Fales) (09/07/89)
In article <4137@csd4.csd.uwm.edu>, peter@csd4.csd.uwm.edu (Peter J Diaz de Leon) writes: > > We have at work some ATT 3B1's and a 20MHz ATT 60386 box. > Both are running ATT Sysyem V. Is there any way to get > a floppy that was formated on a 3B1 read on a 60386 > and vise versa. There is some compatibility, but it is limited. First of all, there is no way to read a mountable disk from one system on the other, the file systems are just too different, so you are limited to cpio. Second, the 6386 does not read the 3b1 VHB, so the only disk format that the 6386 can read is the 8 sector/track, no loader format. (i.e. the disk was created on the 3b1 with fdfmt.nl). The easier direction is to go from the 3b1 to the 6386. The disk is created on the 3b1 using "fdfmt.nl" then written to using "cpio -ocv > /dev/rfp021" Then you can read it back on the 6386 using "cpio -icv < /dev/rdsk/f05d8u" (or f15d8u for drive 1). You can use the -B option on cpio if the archive does not span more than one disk, but if it takes multiple disks you can not use -B. The problem is that when the 3b1 writes the last 5120 byte block on the first disk, and it doesn't fit, it writes the entire 5120 bytes on the next disk. The 6386, on the other hand, expects to see part of the block on the first disk and the remainder on the second. It is slightly more difficult to go the other way, because you have to make sure that there is a VHB on the 3b1 disk. One way is to format on the 3b1, then write the disk on the 6386, finally reading it back in on the 3b1. Alternatively, you can keep on copy of the 3b1 boot track on the 6386 and write it out when necessary. This would be the procedure: #insert a floppy created on the 3b1 with fdfmt.nl into the 6386 dd if=/dev/rdsk/f05d8t of=7300.loader count=8 #insert the new floppy into the 6386, and format if necessary format -vV /dev/rdsk/f05d8t #write the VHB and loadee onto the new disk dd if=7300.loader of=/dev/rdsk/f05d8t #put files on the disk find xxxxx -print |cpio -ocvB > /dev/rdsk/f05d8u (Move to the 3b1) cpio -icvB < /dev/rfp021 -- Peter Fales AT&T, Room 5B-420 2000 N. Naperville Rd. UUCP: ...att!peter.fales Naperville, IL 60566 Domain: peter.fales@att.com work: (312) 979-8031
psfales@cbnewsc.ATT.COM (Peter Fales) (09/08/89)
In article <2986@cbnewsc.ATT.COM>, psfales@cbnewsc.ATT.COM (Peter Fales) writes: > > It is slightly more difficult to go the other way, because > you have to make sure that there is a VHB on the 3b1 disk. One > way is to format on the 3b1, then write the disk on the 6386, > finally reading it back in on the 3b1. Alternatively, you can > keep on copy of the 3b1 boot track on the 6386 and write it out > when necessary. It has been pointed out to me that I forgot to mention the warnings about interchanging disks between 40 and 80 track disk drives. In particular my comment about formatting on the 3b1, writing on the 6386, and reading on the 3b1 is risky (though it may work). The rules are these: - 40 track drives have wide heads. 80 track drives have narrow heads. - Anything written with a wide head can be read (with high confidence) by either a narrow or a wide head. - Anything written by a narrow head on a new (out of the box) or bulk-erased disk can likely be read by either a wide or narrow head. - Anything written by a narrow head down the middle of a wide track previously written on a 40 track drive can likely be read by the narrow head. But if read on a 40 track drive, the head will pick up both the original data and the new data written down the middle. This may cause problems, though in practice you can sometimes get away with it. Summary: If you are just trying to transfer data from machine to another, and the data on the floppy is not critical you may not need to worry. If your only copy of the data is on the floppy, writing on both 40 and 80 track drives is risky at best. -- Peter Fales AT&T, Room 5B-420 2000 N. Naperville Rd. UUCP: ...att!peter.fales Naperville, IL 60566 Domain: peter.fales@att.com work: (312) 979-8031
lenny@icus.islp.ny.us (Lenny Tropiano) (09/08/89)
In article <2986@cbnewsc.ATT.COM> psfales@cbnewsc.ATT.COM (Peter Fales) writes: |>In article <4137@csd4.csd.uwm.edu>, peter@csd4.csd.uwm.edu (Peter J Diaz de Leon) writes: |>> |>> We have at work some ATT 3B1's and a 20MHz ATT 60386 box. |>> Both are running ATT Sysyem V. Is there any way to get |>> a floppy that was formated on a 3B1 read on a 60386 |>> and vise versa. |> ... |>You can use the -B option on cpio if the archive does not span more |>than one disk, but if it takes multiple disks you can not use -B. |>The problem is that when the 3b1 writes the last 5120 byte block |>on the first disk, and it doesn't fit, it writes the entire 5120 |>bytes on the next disk. The 6386, on the other hand, expects to see |>part of the block on the first disk and the remainder on the second. |> ... If I remember correctly, the last time I looked on my 6386E WGS in the office, I vaguely remember a program called /etc/.cpiopc ... Looking at the string table for that (with strings /etc/.cpiopc) I noticed some strikingly familiar strings. Things like: To Exit - press <E> followed by <RETURN>. To continue - insert floppy disk #%d and press the <RETURN> key. ... Just to be sure, I looked at /bin/cpio on that machine and it didn't have the same stuff. So you might want to try that to read floppies that were written with the -B option on the 3B1. Note that /etc/.cpiopc doesn't have the options like: Usage: cpio -o[acvBT#O#] <name-list >collection ^^^^ For buffering and offset location when using the tape device, not to mention that you can't use it with the floppy drive as well... It would be strange (as well as insane) to have multiple savesets on one floppy :-) One aside... that cpio has the options in the usage for T and O, although the Tape documentation recommends using tapecpio (which has those options too, but they aren't listed in the usage). /bin/cpio works just fine with tapes as well... I presumed Thanks Pete for the information on transfering files from the 386 to the 3B1... -Lenny -- Lenny Tropiano ICUS Software Systems [w] +1 (516) 589-7930 lenny@icus.islp.ny.us Telex; 154232428 ICUS [h] +1 (516) 968-8576 {ames,pacbell,decuac,hombre,talcott,sbcs}!icus!lenny attmail!icus!lenny ICUS Software Systems -- PO Box 1; Islip Terrace, NY 11752
hjespers@attcan.UUCP (Hans Jespersen) (09/08/89)
In article <954@icus.islp.ny.us> lenny@icus.islp.ny.us (Lenny Tropiano) writes: >... >If I remember correctly, the last time I looked on my 6386E WGS in the >office, I vaguely remember a program called /etc/.cpiopc ... Looking >at the string table for that (with strings /etc/.cpiopc) I noticed some >strikingly familiar strings. Cpiopc is provided on 386 UNIX for backward compatibility with the 268 UNIX AT&T sold for the PC 6300+. It is not realated in any way (that I know of) to the 3B1. Based on the general soundness of the 286 port ;-) I know I wouldn't trust it for any of my critical data. -- Hans Jespersen UUCP: uunet!attcan!hjespers AT&T Canada Inc. Toronto, Ontario "Yabba Dabba Doo" -- F. Flintstone
les@chinet.chi.il.us (Leslie Mikesell) (09/10/89)
In article <2986@cbnewsc.ATT.COM> psfales@cbnewsc.ATT.COM (Peter Fales) writes:
[good description of using 8-sector format and cpio]
I find it easier to use the "mtools" package on both machines using a DOS
formatted disk. To avoid modifying the filenames by converting to DOS
standards or to compress the data, either zoo or cpio | compress can be
used to generate an intermediate file.
Les Mikesell