Exabyte whiners and real tape drives (tape drive dick length)
Majdi Abbas
So here I am, sitting with *bootable* install media for my VAX[1].
The catch: It's on TK-50 tape. *One* TK-50 tape. Which means I've
got one shot, and only a 33% chance to make it close enough to even get that
shot. That's on a good day. Today is a Monday.
What follows is *not* useful information. If you have one of these,
or have to work with one of these, you're too far gone to be recovering, and
this isn't going to help you any. May the God of DEC have mercy on your
soul.
Oh, and before I get going, those of you who whine about Exabyte
drives not ejecting tapes have no concept of a TK50. TK50's *do not* eject
tapes. You have to arm-wrestle the drive for the tape most of the time, and
even if you're lucky it's a manual eject[2].
Before you can eject the tape, the VAX has to think it's done with the
tape. This is a pretty simple concept, if it worked. The TK50s were not in
production very long, and for good reason. Rumor has it their replacements
are better.[3]
I'm going to describe the operation of a TK-50, ignoring some of the
things that go along with booting a VAX. If you know them, I'm sorry, if
you don't, well, consider yourself extremely lucky.
1) Wait for green light.
2) Pull drive flap up.
3) Stick tape in, right side first or it won't fit.
4) Coerce tape into fitting into the drive.
5) Shove it all the way back
6) Push drive flap down. You may need a hammer.
7) Press Big Red Button. Green light will go off, red light will
start flashing, then go solid.
8) Tape drive begins reading tape.
Here's where we go off onto a tangent for a little bit, although it
is related. TK-50 (drive;cartridge;whole shebang) were designed by complete
absolute fscking lusers.
The cartridges are nice and small. This is because they are just a
reel of tape. The other reel is inside the drive itself. When it starts to
read a tape, it snags the beginning of the tape using a leader that whips
around the spindle of the inside reel, and drags the tape in past the read
and erase heads[4].
9) Drive reads tape, system boots, all is good. </SARCASM>
10) You press the big red button again, wait 45 minutes for the thing
to rewind, then it stops and the green light goes on, it moves a
servo that allows you to move the drive flap again, then you are
permitted to remove the tape. No eject mechanisim whatsoever.
Now, on with the show.
So here I am, booting the VAX.
Things are going good, we get past the 5 minute POST, and the drive
starts *reading* the tape. So the media is good and I'm actually thinking
I have a chance yet. Then the gods decide that they've had enough fun at my
expense, and it's time to get serious.
Loading system software.
2..
?4B CTRLERR, MUA0
?06 HLT INST
PC = 00000E0A
Failure.
>>>
My reaction: "Shit."
VAX's reaction: "Yadda yadda yadda *WHOMP* *SCREEEECCCH* *thwap*"
"THWAPthwapthwapthwapthwapthwapthwapthwapthwap"
[continues]
My reaction: "MotherfuckingasslickingpieceofshitasspirateDECtapedrive."
Actually, my reaction was much more lengthy and probably much more
obscene, but in the afterglow right now that's all I can remember.
Knowing that the tape is hopelessly fucked and there went my last
chance for a while, I don't even bother with the normal procedures. This is
a TK-50. One must adapt constantly or get sucked in.
I quickly wrestle the vax for the tape, remove it, and all seems good.
But I know what's coming. A few hundred feet of half-inch tape, all spooled
into the drives internals.
I spend the next half hour removing tape from the drive, clean it up,
check everything out, decide to try out this unlabeled TK50 I have. Nope,
won't boot. Okay VAX, rewind tape.
[Pause for one hour]
VAX, surely you must be done with it now.
VAX: Nope, it's still in there. I swear.
Me: BULLSHIT. I can hear you flapping around empty. You're flapping
around so much that the VAX is about ready to take off and my hair is being
blown back.
Me: Hits the power switch, pops the thing open, pulls the drive out.
Grabs toolkit and commences disassembly of the drive. Sure enough, it's done.
But I can't get the VAX to let go of the tape until it realizes that it's done,
which isn't going to happen. Powercycling et al will not make it realize that
it's done, it has to feel like relinquishing it's dinner.
So, I'm now dissassembling the TK-50. Sure enough, the magic little
leader that feeds these tapes in is broken. Surprise surprise. Tape looks
okay tho.
It's a couple of hours later, and I have one reassembled TK50, one
sliced hand, a screwdriver with a broken tip, a spare black plastic part,
three spare washers, a couple of spare screws and a spring. The tape is
still in the drive, and I've managed to get all of the first tape into a box
for convenient disposal at my leisure. Like I have leisure. Anyhow, I'd
like to make you an offer: Free TK-50 tape drive, including install media in
need of a manual rewind and a preloaded blank tape. Donatee must pay shipping
and psychiatric admission fee. Includes spare parts[5].
WTB: One SCSI Qbus card.
I am *not* going through this again. Especially because I took
pictures of the aftermath of the first tape, and I'm going to post those
near the VAX as a reminder. Let me know if anyone wants scans.
Every single bad thing you've ever heard about any tape drive doesn't
even begin to describe what the TK-50 is like. Exabyte 8200s have nothing on
these things and never will. DEC was fucking up hardware design years before
the advent of the 8200. I personally believe that the TK-50 is probably what
nearly bankrupted DEC. The number of man-hours wasted in-house wrestling with
these things alone is in the millions. It would have to be.
I saw an RU-81A today. Now *there* is a sight. Appropriately in a
junkyard. For those of you involved with BOFHnet, what do you think of a
bofh.tdfh.tk50? This drive definitely has the FH aspect down.
*sigh*
Down,
not
across
--Majdi
[1] Currently hopelessly crippled due to a drive failure.
[2] The problem is that DEC assumed that the VAX knew more about what was
going on in this drive than the person feeding it ferrite. Boy did
they guess wrong.
[3] They *CANNOT* get any worse.
[4] Which are opposed from each other.
[5] You cannot disassemble one of these drives and not wind up with
spare parts. They can't be anything important, because the drive has
to work to suffer some performance degredation and they don't work, so
there is nothing to degrade.
--
Majdi Abbas <mab...@uiuc.edu> I do not speak for my employer.
"Damn, she looked a lot cuter in the bar..." -- Chris Rioux
(He may be one of my coworkers, but he doesn't speak for them either)
Paul S. Sawyer
In article <slrn5mvi1v...@ux1.cso.uiuc.edu> mab...@uiuc.edu writes:
>
> [...] Rumor has it their replacements
>are better.[3]
>
>[3] They *CANNOT* get any worse.
NEVER say this about a piece of hardware. The designers take it as a
challenge.... Though the TK50 WAS the worst that I have had to deal with.
--
Paul S. Sawyer Paul....@UNH.edu
UNH Telecommunications and Technical Services Voice: +1 603 862 3262
50 College Road FAX: +1 603 862 4545
Durham, New Hampshire 03824-3523
Bill Bradford
Canadian <chr...@uniserve.com> wrote:
: mab...@staff.uiuc.edu (Majdi Abbas) wrote:
: > I'm going to describe the operation of a TK-50, ignoring some of the
: >things that go along with booting a VAX. If you know them, I'm sorry, if
: >you don't, well, consider yourself extremely lucky.
: a bar fridge and serves no purpose other than (a) a tape drive[2] and (b)
: a coffee table?
Hey... I was *given* one a week ago, after posting to one of the workstation
newsgroups that I was looking for a MicroVAX to play around with. This
company said "hey, we've got an old VAX, we'll give it to you, and even
ship it, just to get rid of it but we dont want to scrap it". They
*OVERNIGHT FEDEXED* a box of boards, a case of LN03 laser printer maintenance
kits, two HUGE SMD hard drives, rackmount for said drives, and the MicroVAX
II.
Unfortunately, their idea of shipping the uVAX was "bubble-wrap the chassis
and slap a label on top of the box." When the machine arrived here, it was
in not too good of a shape (chassis beat to shit), which prompted my boss
to say "what the HELL is that piece of crap?"
I managed to give the SMD drives to our tech support crew, which had great
joy in dismantling them (they were non-functional . but damn... works of
art... TONS of machined steel and aluminum). Stripped all the (working)
boards out of the uVAX, along with the TK50 (now in my desk drawer), and
gave the stripped chassis (only power supply and fans left) to a local
high school science academy that needed enclosures for the external hard
drives on one of their servers...
I was supposed to get a *working*, *pristine* MicroVAX II (I've seen it
boot) from another place here in town later that week, in exchange for
doing some minor diagnostic/setup work for them, but I havent heard back
from them lately. So, I'm stuck with a large box full of Qbus boards
until something comes up.
Anybody wanna sell me a uVAX II, VAXstation 2000, or the like? I wanna
get VMS running at home.... yes, I'm a sadist[1]
--
+------------------+-----------------------------------------------------+
| Bill Bradford | Senior Systems Engineer, Sun Microsystems CSA, BOFH |
| mrb...@texas.net | Texas Networking, Inc. http://www.texas.net |
+------------------+--------------------------------+--------------------+
| "I dress in black to mourn the death of colour." | We are a hedge. |
| - the Legendary Pink Dots | Please move along. |
+---------------------------------------------------+--------------------+
[1] I "grew up" on a VAX 3000 running VMS 5.5-2, before I ever touched my
first UNIX system . . .
Bill Bradford
Bill Bradford <mrb...@texas.net> wrote:: : > I'm going to describe the operation of a TK-50, ignoring some of the
: : >things that go along with booting a VAX. If you know them, I'm sorry, if
: : >you don't, well, consider yourself extremely lucky.
: : a bar fridge and serves no purpose other than (a) a tape drive[2] and (b)
: : a coffee table?
: Hey... I was *given* one a week ago, after posting to one of the workstation
: newsgroups that I was looking for a MicroVAX to play around with. This
: company said "hey, we've got an old VAX, we'll give it to you, and even
: ship it, just to get rid of it but we dont want to scrap it". They
: *OVERNIGHT FEDEXED* a box of boards, a case of LN03 laser printer maintenance
: kits, two HUGE SMD hard drives, rackmount for said drives, and the MicroVAX
: II.
Did I mention that said company paid $800 for said shipping? To *give*
me a system that wasnt worth $200, maybe?
I have a feeling someone in their shipping department was on drugs...
Christer Weinigel
In article <slrn5mvi1v...@ux1.cso.uiuc.edu>,
Majdi Abbas <mab...@uiuc.edu> wrote:
[rant about the luser nature of TK50:s snipped]
Hmm, I rather like the beasts, they've been rather nice to me, all you need
is a bit of patience and a sense of fatalism when it eats your tape [1].
I got my newsfeed by TK50[2] once upon a time and it worked quite well
most of the time, I've had much more trouble with DAT:s and Exabytes.
[1] Just once actually. BTW. There is a _lot_ of tape in a cartridge.
[2] Sneakernet mark II.
--
If it's tourist season, why can't we shoot them?
Mark Stapleton
Bill Bradford <mrb...@texas.net> wrote:
>Did I mention that said company paid $800 for said shipping? To *give*
>me a system that wasnt worth $200, maybe?
>
>I have a feeling someone in their shipping department was on drugs...
Possibly. However, I suspect an ulterior motive. Scrapping those
puppies in a politically correct manner requires stupendous expenses
to comply with EPA *toxic waste* requirements.
'For Finagle's sake, why?' I hear you cry. Probably a trace of cadmium
or other heavy metal in components. Just enough to give a government
bureaucrat a case of the shingles, but not enough to actually *do* any
harm.
--
"Bowling fulfills the human need for multi-toned shoes and beer frames!"
Mark Stapleton (mst...@swbell.net)
Petro
In article <3371fe03...@snews5.zippo.com>,
Mark Stapleton <mst...@swbell.net> wrote:
>Possibly. However, I suspect an ulterior motive. Scrapping those
>puppies in a politically correct manner requires stupendous expenses
>to comply with EPA *toxic waste* requirements.
>'For Finagle's sake, why?' I hear you cry. Probably a trace of cadmium
>or other heavy metal in components. Just enough to give a government
>bureaucrat a case of the shingles, but not enough to actually *do* any
>harm.
My father had a lighter that he was particularly fond of, made
by colibri(sp?) that is now discontinued for public saftey reasons.
This particular lighter you ignite by opening the lid (which
startes releasing the butane) then you push the little electric button
to cause a spark.
Seems that some brainless government cretin (is that redundant?)
noticed that there was about 1/2 second where the gas was flowing unimpeded.
I think it's damn near time to raise the black flag.
--
***************** PLEASE TAKE NOTE:
In an effort to reduce the amount of junk mail that I receive, I am no longer
reading email sent to pe...@suba.com. send email to lo...@encodex.com where
login = petro. You send unsoliciated commercial email, and I will kill you.
Frank McConnell
Majdi Abbas <mab...@uiuc.edu> wrote:
>[2] The problem is that DEC assumed that the VAX knew more about what was
> going on in this drive than the person feeding it ferrite. Boy did
> they guess wrong.
Not always.
I worked for an interesting company that seemed to get its backup
operators by "promoting" folks from the shipping department. The
clueful ones with the BOFH nature soon moved to either doing
PC/terminals/datacomm stuff, or moved to doing phones (i.e. moving
phones and taking care of the PBX) -- both of which offered better
opportunities for the clickety-click sort of instant gratification.
Once upon a time, someone who wasn't clueful got assigned to be the
backup operator. This is a story about him.
He was doing the backups one weekend. Backups included three or four
MicroVAX II-class machines with TK50 drives. And they were done to
whatever tapes he had; these generally lived in mail cartons or file
boxes that were left somewhere in the vicinity of the machines being
backed up, as near as I can remember.
I happened to come in that weekend. None of the backups worked, he
said. Uh-huh, I thought. "Can you show me?"
So he took his tape, plugged it into the gaping TK50 maw, pressed
the button, and we were both treated to a loud BRAAAP sound from
Deep Within. Yep, the takeup reel was trying to shred its leader.
I don't remember how, but somehow I managed to stop it from
succeeding. This machine may have been one of the ones that was
missing some of its skin, and if so I probably just pulled the TK50
drive out so I could disconnect the power at the drive. If not I
turned the VAX off (shrug).
Well, anyway I got the tape out, and found that the little loop wasn't
a closed oval anymore -- the tip was missing, and the takeup reel leader
would slip right through the opening.
Come to find out that Mr. Backup Operator, upon each of the previous
failures, somehow got the damnable TK50 drive to give the tape back so
he could use it in the next system.
OK, so he didn't twig to the notion that maybe the tape was the
problem, even after a couple of failures in different drives and
systems. I can understand that, he was basically clueless. What I
can't figure out is how he got the tape back out of the drive.
...
I guess what it all comes down to is, Cartridge Tapes Suck. TK50s,
the oddball HP/3M Things That Look Like QIC, Exabytes, DDS....
It kind of makes me want to use nine-tracks. They suck too, but at
least you can keep an eye on them.
-Frank McConnell
Don Stokes
Frank McConnell <f...@aphasia.us.com> wrote:
>It kind of makes me want to use nine-tracks. They suck too, but at
>least you can keep an eye on them.
Only if they're the vacuum column variety, in which case they boh suck
and blow. The non-vacuum variety just blows. (Unless it's an auto-loader.)
--
Don Stokes, Network Manager, NetLink, PO Box 1762, Wellington, New Zealand
d...@netlink.co.nz (work) d...@zl2tnm.gen.nz (home) Phone +64 4 495-5052
Joe Zeff
In an egregious waste of bandwidth, Chris King
<ch...@csking.demon.co.uk> wrote:
>DEC themselves are just as bad. How's about this - they once sent out
>several packages by motorcycle courier, individually wrapped, which
>appeared to be PAK's
I was reading The RingWorld Throne over the weekend, and read that as
implying Pak Protectors. The BOFH's worst nightmare. Or, the BOFH to
end all BOFHs. Take your pick.
------------------------------------------------------------
Joe Zeff Earthlink Network
jo...@earthlink.net Senior Support JOAT
(800) 395-8410
"The only problem with troubleshooting is that
sometimes trouble shoots back."
------------------------------------------------------------
bert hubert
On Mon, 12 May 1997 20:11:32 GMT, Joe Zeff <jo...@earthlink.net> wrote:
>I was reading The RingWorld Throne over the weekend, and read that as
^^^^^^^^^^^^^^^^^^^^
Is it any good? The first two were excellent..
--
Delft University of Technology, department of Physics
Phone: +31-15-2786122 / Lorentzweg 1, 2628 CJ, Delft, The Netherlands
Inspice et cautus eris - D11T'95
Peter Gutmann
jo...@earthlink.net (Joe Zeff) writes:
>In an egregious waste of bandwidth, Chris King
><ch...@csking.demon.co.uk> wrote:
>>DEC themselves are just as bad. How's about this - they once sent out
>>several packages by motorcycle courier, individually wrapped, which
>>appeared to be PAK's
>I was reading The RingWorld Throne over the weekend, and read that as
>implying Pak Protectors. The BOFH's worst nightmare. Or, the BOFH to
>end all BOFHs. Take your pick.
Of course they're not Pak Protectors! Everyone knows they're German anti-tank
guns, probably intended to keep the lusers at bay (you'd need a PAK to
penetrate the skulls).
Peter.
Joe Zeff
In an egregious waste of bandwidth, a...@vvtp.tn.tudelft.nl (bert
hubert) wrote:
>On Mon, 12 May 1997 20:11:32 GMT, Joe Zeff <jo...@earthlink.net> wrote:
>
>>I was reading The RingWorld Throne over the weekend, and read that as
> ^^^^^^^^^^^^^^^^^^^^
>
>Is it any good? The first two were excellent..
>
>
*I* liked it twice. YMMV. Larry told me, "I'm told my books are said
to be re-readable."