• Please review our updated Terms and Rules here

RT-11 transferring binary images to serial console

H

hudson

New Member
Joined
May 20, 2014
Messages
3
Location
NYC


I'm one of the folks who picked up the PDP-11/34 from the NYC craigslist a few weekends ago. After cleaning out the dust and contacts, reseating all the cards and cables, and removing all of the decayed air filters, we have one of the cabinets booting RT-11 from the RK05 drives really well and have restored several of the VT100 terminals. We've written some initial notes on the NYC Resistor blog http://www.nycresistor.com/2014/05/15/pdp-11/ and I'm collecting further notes on my own site http://trmm.net/PDP-11. We haven't powered up the dual cabinet machine, RL02 drives or TU10 tape unit yet and are researching the correct recipes for baking the tapes prior to attempting to read them.



One of our initial goals is to make copies of the various decpacks that were in the lot. It looks like COPY/DEVICE/BINARY RK0: TT: should do it, but both in simh and on the real hardware the NUL characters are not being output. I see that SET LP CTRL will allow the line printer to receive control characters, but there doesn't seem to be an equivalent option for the TT device. As another work around, I can use DUMP/NOASCII/WORDS/TERMINAL RK0:, but this bloats up the data by at least 4x and the terminal link is only 9600 baud so each 2.5MB pack would take nearly two hours. Is there an easy way to enable a raw output mode with COPY, or do I need to write a something that does bit-stuffing or base64 encoding?
 
I'm curious to hear about this - the DUMP option was the one I ended up taking to back up my 11's hard drive when I got it, but it's certainly not an optimal solution.

P.S.: Sweeet system.
 
If the system has another serial port and you can boot a an RT-11 installation that includes the TU58 DECtape-II driver, then I suggest dumping to an emulated TU58 tape. Real TU58 tapes hold a mere 256k, but the protocol allows for up to 65535 blocks (one block less than 32 megabytes, and the maximum RT-11 volume size). The TU58 driver needs a simple patch to support the larger volume sizes, and folks here can help you with that if you need. The bit of unintuitive magic that you may or may not already know is that TU58 tapes are treated like random access block devices, and formatted with RT-11 filesystems just like disks. Thus, they're essentially slow disk drives that work over ordinary serial ports, making them very handy for bootstrapping and archival tasks like you are working on.

The AK6DN tu58em program works well for this, although slowly since everything's going over a slow serial link.

I've used this technique to copy floppies on my PDP-11/03 to image files on an emulated TU58 volume, with tu58em running on my Macbook.

It may take a bit of jumping through hoops to get it working, but we can help. You have some cool hardware there, and I'm happy that you're treating it right!
 


We temporarily disconnected the second rack with a RK05, TM10 and TU11 and installed a terminator card in its place. The second PDP-11/34 powers up just fine and both RL01 drives work perfectly. We only seem to have one boot disk, with RT-11S V04.00. The other packs contain source code for RSX-11M and EPACH (?), but when we try to boot with them report ***THIS VOLUME DOES NOT CONTAIN A HARDWARE BOOTABLE SYSTEM ***.

Since we now have the much larger disks running, finding a faster way to retrieve data will be key. An emulated hardware device is looking more attractive than 9600 baud...
 
hudson said:
researching the correct recipes for baking the tapes prior to attempting to read them.
Click to expand...

The critical part is providing adequate (ie. a LOT) of airflow and maintaining a constant temperature. I've processed
over a thousand 1/2" tapes and the system I came up with is a chamber with fans above and below the reels with
a heat source below the lower set of fans. The entire box is a closed recirculating air system with a thermostat.
Temp is around 130F, tapes are hung in the center with a rod, the end of the tape is fastened to the outside of
the hub with a piece of cellophane tape. They are kept in the chamber for about 24 hours. You can make a rough
guess on the stickyness of the tape my how much stick there is to the outer wraps of the reel. The tape shouldn't
adhere too easily to the inner wraps (it's hard to describe but you can tell the difference before and after leaving
a tape in the chamber).

You can see it on the right side of this picture. I had humidity monitoring on it as well.
oven.jpg
 
If you only have one bootable copy of RT-11 make a copy of that disk now! It’s too easy to screw up that one good copy and then you have a real problem. Thought you had several disks that came with that system so you would be well served by building a couple more system disks and putting your one good copy of RT-11 on the shelf.
I have two RL drives on my 34 and it’s a fairly easy process to build copies of the disks from one drive to the other. One of the other things I have taken to doing is keeping a system disk with RT-11 and all the other junk like Basic in drive “0” with the write disable turned on to keep me from overwriting the system disk and use disk “1” as the volume. Do you have a copy of Basic?
 
We have many bootable copies of RT-11 on RK05 decpacks, but only a small number of RL01 platters. And only one of those seems to be bootable. The others are for RSX-11M, but don't boot with the DL boot rom. One is labeled "RSX11M V4.0 ID:MCRSRC RL1 4/5", which makes me wonder if we're missing four others. And yes, BASIC.SAV is on the RT-11 V04.00 RL01 platter.


We also encountered our first failed platter on a decpack last night. The "vent" on the bottom had been pushed upwards enough to dent the platter it seems. Luckily it didn't destory the drive before we noticed.
 
That damaged platter looks scary. I'm going to visually inspect every single one of my RL02 packs before they go in one of my drives!
 
That pack would be a drive trasher for sure. ALWAYS visually inspect any unknown pack before putting it into a good drive.

Hudson, do you have any people onsite helping you who have firsthand experience with this type of gear? I strongly suggest you find a local old-timer to help you navigate without destroying anything by accident. And I don't count as an old-timer (but fortunately I can ring my father to discuss anything before I destroy it.)

Lou
 
RetroHacker_ and I now disassemble every pack, clean, and inspect before attempting to mount. We've found some really dirty packs that turned out to be fine (just a little dust inside), and we've found excellent looking packs that, after inspection, were revealed to be heavily damaged and likely would have destroyed our drive(s).

vtserver is your best bet for making 1:1 emulator-bootable copies of DEC media. I believe we've been in contact about it previously. I'm working on a modification to DL11/DLV11 serial line unit boards to increase speed potentially as high as 2.3 Mbit/sec for media archive and restore. That would reduce archive times for RL02 packs (10 MB) to a few minutes.
 
You must log in or register to reply here.
Back
Top