• Please review our updated Terms and Rules here

Archive Corp Model 2060E QIC-02

Hopefully I don't annoy anyone by bringing up the nominal topic of the thread again, but I received the box of plastibands today from the bezoid. There were 5 green bands (I haven't counted the others). I put one of them on the cartridge where the tension band had broken and was able to read about 85% of the data before I got to a spot on a track where, presumably, the crinkle from the tape salad was too severe to read. In the same general zone on earlier tracks, the drive would shoe-shine a bit trying to get a good read and succeed occasionally, but the 83% error seems persistent. Again, this was a redundant tape, so no data was lost, but it valiantly gave its life for science. I also have a box of the misumi bands on the way: https://us.misumi-ec.com/vona2/detail/223000860382/?ProductCode=MB-15063WA-100G which I think I will wait for before trying the SYTOS tapes.
 
I lied. I didn't wait. I transferred the plastiband from my experimental/sacrificial cartridge to the first (of 7) of the SYTOS backups from 1992. But I just used dd to get a dump. Like I observed with the DDS tapes written with SYTOS, there was an initial short tapefile (a single block of 1024 bytes) followed by a long tapefile (61,972 blocks of 1k each, a little more than expected, I presume because these cartridges are more than the nominal 600 feet). I don't have any real idea of how SYTOS organizes its backup data, but "strings" showed me the kinds of things I'd expect. One interesting new failure mode: When I went to eject the tape, the swiveled out tape cover door thingie did not return to its "covered" position and the cartridge didn't come out of the drive slot. Fortunately, I have the drive out of the case and all the parts are visible, so I was able to see what was wrong and could manually close the cover and the tape then ejected fine. Maybe the corner hinge needs some lubricant or maybe when I re-closed the tape cartridge after my tension belt surgery, something was pinched or I lost a spring or something? Not sure, but I don't recall reading about this hazard elsewhere, so I thought I'd mention it. Now I'll wait for the misumi bands before continuing.
 
There's a very small round spring that lies on the shaft between between the "door" and the baseplate. One end of said spring engages a small slot in the plastic "door" pivot; the other lower end is retained by pressing against the base of the shaft.
It's quite easy to disengage one end or the other of the spring such that the door has no spring tension to close it. Getting it reset can be a bit frustrating, but can be done. This mechanism is present, as far I can tell on every size of DC and Travan cartridge.
 
I checked the pivot, it seems the plastic case was not quite clicked all the way down. After removing the plastic case, I noticed that the door sprung closed as seemed normal and reattaching the plastic case and ensuring it was clicked all the way down, the door no longer stuck open. All fixed! Thank your for pointing me in the right direction.

With all the talk about shedding in various threads, I have been looking for any evidence and not finding it. What seem like the primary hazards beyond the tension bands: pinch roller turning to goo and the tape shedding don't seem to have afflicted me. It could be that it's because the drive and tapes have spent most of the last 30 years in a fairly temperature stable and low humidity environment. Or maybe I just "live right". I guess we'll see for sure when the misumi bands arrive in a week or two, and I take a go at SYTOS. As far as I can tell, I have a DOS environment reconstructed that is very close to identical to what I used to create them.
 
Shed on the tape will accumulate on the two pins attached to the baseplate
that the tape passes across.
If the buildup is severe enough, the tape will stick when it serpentines and deposit
the crud onto the tape, eventually smearing it out of the reversal zone.
My solution was to make a sacrificial base plate with pads made from lint-free cloth
glued over the posts. Preferably the pads are loose enough they can spin a bit so
the crud is spread across them.
 
Last edited:
Al Kossow said:
Shed on the tape will accumulate on the two pins attached to the baseplate
that the tape passes across.
If the buildup is severe enough, the tape will stick when it serpentines and deposit
the crud onto the tape, eventually smearing it out of the reversal zone.
My solution was to make a sacrificial base plate with pads made from tyvec cloth
glued over the posts. Preferably the pads are loose enough they can spin a bit so
the crud is spread across them.
Click to expand...
I have inspected closely and see no shedding. The pins are spotless.
 
I think I'm outing myself as impatient, but I found another tape to risk while I wait for the misumi bands. It was an Archive Model 560 (600 foot) cartridge that I'd had success with earlier, even without replacing the tension bands. I put a plastiband on it, gave it a retension under linux and then rebooted into my 1993 era, MS-DOS v5 install with the SYTOS software I presume I was using before linux appeared on my scene, circa September 1992. So, I fire up SYPLUS.EXE and get oriented to their bizarre procedures, and setup one to RESTORE from the tape. Weirdly (how could my 30 year old memory be faulty in any way???), SYTOS (Version 1.32 for DOS) says: "Information exists on the media in the backup device. Sytos Plus cannot identify the contents using the currently selected format." WHUT? So, I boot back into linux and try to read it. Initially with dd bs=1024 (which has worked before) I get a failure. But then I think to try bs=512, and I get a single block read for the first tape file. Then without rewinding, I run dd again (on tapefile 2) initially with bs=512, which stopped part way through. Then I rewound, skipped the first tape file with "mt fsf 1" and the dd'd with bs=1024 and got about 50MB before it stopped normally. I did a second read of the second tape file and got exactly the same thing (md5sums are identical). Could this NOT be a SYTOS backup? That seems so weird. Why/how would I have done that? For reference, a hexdump -C of the first 512-byte tapefile looks like this:
00000000 dc c5 c6 bb d9 c2 cb d8 c5 c4 d5 ba b3 b8 be b3 |................|
00000010 b8 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a |................|
00000020 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a |................|
*
00000040 8a 8a 8a 8a b3 b8 ba b3 b8 be ba b3 ba b3 bb bf |................|
00000050 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a 8a |................|
00000060 8a 8a 8a 8a 8b 8a 1f 00 45 43 43 5e 33 30 30 00 |........ECC^300.|
00000070 00 00 00 00 03 56 1e 05 42 52 45 41 4b 02 03 2f |.....V..BREAK../|
00000080 06 56 45 52 49 46 59 02 46 2f 03 53 45 54 02 e6 |.VERIFY.F/.SET..|
00000090 1a 06 50 52 4f 4d 50 54 02 cc 1a 04 50 41 54 48 |..PROMPT....PATH|
000000a0 02 3c 17 04 45 58 49 54 00 35 1a 04 43 54 54 59 |.<..EXIT.5..CTTY|
000000b0 03 77 18 04 45 43 48 4f 02 c9 2e 04 47 4f 54 4f |.w..ECHO....GOTO|
000000c0 02 cc 0b 05 53 48 49 46 54 02 51 0b 02 49 46 02 |....SHIFT.Q..IF.|
000000d0 ff 09 03 46 4f 52 02 42 0e 03 43 4c 53 00 e8 17 |...FOR.B..CLS...|
000000e0 08 54 52 55 45 4e 41 4d 45 03 b3 19 00 2e 43 4f |.TRUENAME.....CO|
000000f0 4d 2e 45 58 45 2e 42 41 54 56 42 41 50 57 01 00 |M.EXE.BATVBAPW..|
00000100 02 00 80 02 43 50 20 00 00 00 00 02 01 00 55 6d |....CP .......Um|
00000110 88 5c 00 01 03 01 00 55 6d 88 5c 00 01 02 01 00 |.\.....Um.\.....|
00000120 55 6d 88 5c 00 00 00 02 00 55 6d 88 5c 01 2f 50 |Um.\.....Um.\./P|
00000130 00 b4 5c 00 00 01 ba 5c 00 00 01 20 02 00 55 6d |..\....\... ..Um|
00000140 c3 5c 00 03 00 00 02 00 cd 5c 66 d0 5c 4f 4e 00 |.\.......\f.\ON.|
00000150 4f 46 46 00 d7 5c 00 00 01 dd 5c 00 00 01 80 00 |OFF..\....\.....|
00000160 00 55 6d e6 5c 00 01 01 01 64 00 00 00 e7 03 00 |.Um.\....d......|
00000170 00 00 00 f6 5c 00 00 01 fc 5c 00 00 01 10 00 00 |....\....\......|
00000180 5d 6d 88 5c 00 08 5d 00 00 01 0e 5d 00 00 01 08 |]m.\..]....]....|
00000190 00 00 65 6d 88 5c 00 1a 5d 00 00 01 20 5d 00 00 |..em.\..]... ]..|
000001a0 01 01 01 00 6d 6d 88 5c 00 2c 5d 00 01 01 8a 5c |....mm.\.,]....\|
000001b0 00 00 35 5d 00 00 01 93 5c 00 00 3e 5d 00 01 01 |..5]....\..>]...|
000001c0 8a 5c 01 a5 5c 00 49 5d 00 00 01 9c 5c 02 a5 5c |.\..\.I]....\..\|
000001d0 53 5d 00 00 00 02 00 55 6d 88 5c 01 2f 57 00 62 |S].....Um.\./W.b|
000001e0 5d 00 02 02 8a 5c 8a 5c 00 00 6d 5d 00 01 01 73 |]....\.\..m]...s|
000001f0 5d 00 00 00 20 11 00 55 6d 88 5c 00 00 00 00 00 |]... ..Um.\.....|
00000200
Click to expand...
Anyone happen to recognize that?
 
Strip the high-order bit from the first 16 or so bytes and you get:

dc c5 c6 bb d9 c2 cb d8 c5 c4 d5 ba b3 b8 be b3 b8
5c 45 46 3b 59 42 4b 58 45 44 55 3a 33 38 3e 33 38
\ E F ; Y B K X E D U : 3 8 > 3 8 followed by lots of linefeeds

Doesn't look like a sytos backup to me, unless the data is very corrupted.
 
Yeah, I noticed the high bits too, and was excited to try to see what it might be spelling with the high bit stripped. And then at offset 0x66, the high bit stops, mostly. There are several tapes from the same epoch and the "header" files all look similar. I can diff the the hexdumps and the differences are in isolated spots. My sense is that if it was corrupted, the drive would be complaining about bad checksums and rereading things. The body (tapefile 2, which starts similarly) streams continuously. I know I had SYTOS and used it, so I wonder what else I could possibly have been using. I do have notebooks from the era, I guess I better go back and read through them to look for clues. SYTOS seems to be set up for use with the Archive drive and the DDS-1 drive, they are both listed in stored configurations in my July 1993 DOS restoration.
 
Welcome to my world ("I've got this box of tapes and I can't tell you what the host system was or what was used to write them....can you get the files for me?")
Sometimes, if it's a full system backup, the name of the backup program will be found in a directory. Just your basic sleuthing.
 
Dunno--one of the reasons that I roll my own controllers. Anything you can find is likely 30+ years old and technology has come a long way. Modern medium-scale MCUs have lots of timers and DMA channels in addition to being very fast and having lots of memory.

Anent that, there's a Fuji M2444 drive and Laguna controller in Reno for cheap, apparently: https://reno-nv.americanlisted.com/...u-sun-m2444-tape-drive-computer_23183767.html Bring your own truss...
 
Chuck(G) said:
Dunno--one of the reasons that I roll my own controllers. Anything you can find is likely 30+ years old and technology has come a long way. Modern medium-scale MCUs have lots of timers and DMA channels in addition to being very fast and having lots of memory.

Anent that, there's a Fuji M2444 drive and Laguna controller in Reno for cheap, apparently: https://reno-nv.americanlisted.com/...u-sun-m2444-tape-drive-computer_23183767.html Bring your own truss...
Click to expand...
That is the combo I have for 9-track. I don't actually have any data on 9-track tapes though. We used it for data transfer only to/from dinosaur collaborators. I also have a never tested front-loading 9-track drive alleged to have a SCSI-Pertec adapter.
 
The favorite drive of my career was the CDC 607 (in my avatar) 1500 lbs of a built-like-a-battleship drive. Vacuum counter-rotating capstans. Very clever design. The 65x's were the cost-reduced blue-glass version, with a much weaker voice-coil valve.

I prefer the front-loading drives for my work--I can watch what's going on. Was working with a customer's tape a couple of nights ago. Terrible condition, when the drive suddenly dropped "ready". I had a terrible vision of having to troubleshoot the drive. When coaxing the drive back to operation, I could see (viewing from above) that there were feet of nice clear tape base. Drive figured that it was dealing with a broken tape.
 
Last edited:
russell-- said:
[...] I know I had SYTOS and used it, so I wonder what else I could possibly have been using. I do have notebooks from the era, I guess I better go back and read through them to look for clues. SYTOS seems to be set up for use with the Archive drive and the DDS-1 drive, they are both listed in stored configurations in my July 1993 DOS restoration.
Click to expand...
I reviewed my notes from the time of the backup, September 1992, and there is nothing in them that would indicate I was using anything but SYTOS on the QIC drive in that period. A few weeks later, I'm talking about buying the first DDS drive.
 
That header doesn't look like it's from a Sytos backup. Could it be a continuation of another volume? If it's not too huge, you can post it privately on a cloud server and I'll have a look at it.
 
Just to follow up, after some cognitive assistance from @Chuck(G), I managed to piece together what Sytos Plus was trying to tell me, which I can paraphrase as: "there are multiple possible formats and we aren't going to try to detect which one it is, you have to choose from a list of two", so I tried "the other one" and that detected data as Sy-tos 3.x (not Sytos Plus 1.0) and proceeded to add the volume and is currently recovering files from it. Also, @Chuck(G) pointed me at a tool for decoding Sytos tape dumps: http://thornton.info/tools/010editor/SytosPlus.bt which might have proved useful.
 

Attachments

  • PXL_20231110_085153463.jpg
    PXL_20231110_085153463.jpg
    5.7 MB · Views: 7
Last edited:
You must log in or register to reply here.
Back
Top