[comp.sys.3b1] Floppy problem

zaft@suned1.nswses.navy.mil (Gordon C Zaft) (03/04/91)

	Well, I was too smart by half for my britches.  My newly arrived 
7300 was working nicely, but I decided to replace the floppy drive since
(as I'd been warned) the floppy drive was screwy (it would read floppies
but not write/format).

	I went out and bought a nice, brand-new Teac 55BR; changed the
jumper for drive select to DS0, and put it in the machine.  Now the
drive won't format, write, OR read!  Any brilliant ideas?  There's
only one line of jumpers I can see, and the only jumper that made
sense to me was the drive select.  The old drive had oodles of jumpers
on it, but then, it had a lot more chips on  it too.

	As if that wasn't bad enough, at the same time (I THINK) as I
changed the floppy drive, the machine starting to do the double-boot
thing on powerup; it would boot, get to "checking stored files", then
reboot and be fine.  If I rebooted once the machine had been up,
it was fine.  Is this a power supply problem or is it related to the
floppy thing?

	This is a 1 MB motherboard; I'm running 3.51 (no fixdisk yet,
I wanted to be able to save it floppy before installing it!).  The 
old floppy was a  Teac 55B.  I also have a 0.5M RAM card in the 
machine.

	Thanks in advance.
--
+  Gordon Zaft                        |  zaft@suned1.nswses.navy.mil         +
+  NSWSES, Code 4Y33                  |  suned1!zaft@elroy.jpl.nasa.gov      +
+  Port Hueneme, CA 93043-5007        |  Phone: (805) 982-0684 FAX: 982-8768 +
** Ray Kaplan is a God! *** "Quoth the raven: Eat My Shorts!" -- B. Simpson **

floyd@ims.alaska.edu (Floyd Davidson) (03/04/91)

zaft@suned1.nswses.navy.mil (Gordon C Zaft) writes:
[...]
>7300 was working nicely, but I decided to replace the floppy drive since
>(as I'd been warned) the floppy drive was screwy (it would read floppies
>but not write/format).
>
>	I went out and bought a nice, brand-new Teac 55BR; changed the

I can't help on the new drive because I don't know what a 55BR is
compared to a 55B, but...

You might tried cleaning the old one up and putting it back in.
Be careful to clean the little led and detector that shines through
the write protect notch on the left side of the floppy disk.  Chances
are very good that the only problem with the drive is that it thinks
everything is write-protected because of dust on one side or the other
of that led/detector combination.  It has happened to several of us!

>
[...]
>	As if that wasn't bad enough, at the same time (I THINK) as I
>changed the floppy drive, the machine starting to do the double-boot
>thing on powerup; it would boot, get to "checking stored files", then
>reboot and be fine.  If I rebooted once the machine had been up,
>it was fine.  Is this a power supply problem or is it related to the
>floppy thing?
>

Maybe neither.  Once you have the machine up try running "/etc/fsck -D"
and see what you get.  The initialization script, /etc/rc, has a
good/bad idea in it about checking the file system.  Most UNIX systems
don't come up automatically, but the UnixPC was supposed to be user
friendly, so it does.  And it also hides from you everything that
the file system check finds.  There is a place in the /etc/rc script
where it does something like "fsck -pq /dev/fp002 || { more script }",
where the { more script } is a reboot (Mine is long since changed, so
I'm not sure exactly how it is done...)  The 'q' option tells fsck
to be quiet and not tell you what it is doing; the 'p' option tells
fsck to "fix minor problems"  ( see fsck(1m) ).  Running fsck on
/dev/fp002 will cause a reboot if anything is "fixed".

That is where your reboot is coming from, most likely.  It seems
that something is getting fixed everytime, but you can't see what
it is because of the way the script is arranged.

Floyd
-- 
Floyd L. Davidson  |  floyd@ims.alaska.edu   |  Alascom, Inc. pays me
Salcha, AK 99714   |    Univ. of Alaska      |  but not for opinions.

dnichols@ceilidh.beartrack.com (DoN Nichols) (03/05/91)

In article <1991Mar4.130902.24392@ims.alaska.edu> floyd@ims.alaska.edu (Floyd Davidson) writes:
>zaft@suned1.nswses.navy.mil (Gordon C Zaft) writes:
>[...]
>>	As if that wasn't bad enough, at the same time (I THINK) as I
>>changed the floppy drive, the machine starting to do the double-boot
>>thing on powerup; it would boot, get to "checking stored files", then
>>reboot and be fine.  If I rebooted once the machine had been up,
>>it was fine.  Is this a power supply problem or is it related to the
>>floppy thing?
>>
>
>Maybe neither.  Once you have the machine up try running "/etc/fsck -D"

	[...]

>/dev/fp002 will cause a reboot if anything is "fixed".
>
>That is where your reboot is coming from, most likely.  It seems
>that something is getting fixed everytime, but you can't see what
>it is because of the way the script is arranged.

	Are you using /etc/shutdown, or the shutdown entry from the
'install' login?  If not, or if you are not waiting for it to complete, you
are PRODUCING problems for it to fix.  Just shuting down a unix computer is
about the worst thing you can do for file system sanity.  (Except in some
strange beasties, such as the Textronix 6130, and maybe its brethren, which
have what is called a 'soft power switch'.)  When you turn it off, the led in
it changes from steady to flashing, and the system starts a shutdown
sequence.  It runs fsck before the shutdown completes, and sets a file as a
flag that the fsck has been done.  Once this is all complete, it then pulls
the rug out from under itself, and the led on the switch finally goes out.
Boot time is reduced, since it already has a certified filesystem.  (Of
course no one is going to drop-kick the cpu box between the shutdown and the
re-boot, are they :-)

	Sorry if I've pointed out something which you already were fully
cognizant of.  Sometimes I forget simple things, and feel quite stupid when
I discover them.  (Especially if someone else points them out to me :-)

	Good luck with your floppies.
		DoN.

-- 
Donald Nichols (DoN.)		| Voice (Days):	(703) 664-1585
D&D Data			| Voice (Eves):	(703) 938-4564
Disclaimer: from here - None	| Email:     <dnichols@ceilidh.beartrack.com>
	--- Black Holes are where God is dividing by zero ---

slimick@unix.cis.pitt.edu (John C Slimick) (03/05/91)

We have our two 7300's up and running, but in
installing the software, we stumbled on disk 11
of the UNIX utilities: Enhanced Editors V3.5

We are running UNIX 3.0, with the following packages
installed:

GSS-DRIVERS C Language V1.01
GSS DRIVERS V 1.01 Install Set 1
GSS DRIVERS V 1.01 Install Set 2 
GSS Solutions GSS TERMINAL v1.02
Development Set 3.5
Document Preparation Set 3.5
3.0 ar and 3.0 nm
UNIX PC Curses/Terminfo

We have two copies of  the  Enhanced Editors diskette, one
of which was shrink-wrapped when we got it.

We are tryining to install it under the systems administration
Software setup/ Install software from floppy window. We get
"diskette unreadable" or "diskette inserted wrong" (which it isn't).

Any clues?

john slimick
university of pittsburgh at bradford

slimick@unix.cis.pitt.edu

Mariusz@fbits.ttank.com (Mariusz Stanczak) (03/05/91)

In article <1991Mar4.130902.24392@ims.alaska.edu> floyd@ims.alaska.edu (Floyd Davidson) writes:
>That is where your reboot is coming from, most likely.  It seems
>that something is getting fixed everytime, but you can't see what

Also, if there's a "serious" damage discovered by `fsck', the "fixed"
file can be found in /lost+found... if there's anything there (and
usually, on a good file system, it would just be a shell history file),
you can examine it to see whether it's anything to be concerned about.

-Mariusz
-- 
INET: Mariusz@fbits.ttank.com
CIS : 71601.2430@compuserve.com
UUCP: ..!uunet!zardoz!ttank!fbits!Mariusz

floyd@ims.alaska.edu (Floyd Davidson) (03/05/91)

In article <1991Mar5.023004.2855@ceilidh.beartrack.com> dnichols@ceilidh.beartrack.com (DoN Nichols) writes:
>In article <1991Mar4.130902.24392@ims.alaska.edu> floyd@ims.alaska.edu (Floyd Davidson) writes:
>>zaft@suned1.nswses.navy.mil (Gordon C Zaft) writes:
>>[...]
>>>	As if that wasn't bad enough, at the same time (I THINK) as I
>>>changed the floppy drive, the machine starting to do the double-boot
>>>thing on powerup; it would boot, get to "checking stored files", then
>>>reboot and be fine.  If I rebooted once the machine had been up,
>>>it was fine.  Is this a power supply problem or is it related to the
>>>floppy thing?
>>>
>>
>>Maybe neither.  Once you have the machine up try running "/etc/fsck -D"
>
>	[...]
>
>>/dev/fp002 will cause a reboot if anything is "fixed".
>>
>>That is where your reboot is coming from, most likely.  It seems
>>that something is getting fixed everytime, but you can't see what
>>it is because of the way the script is arranged.
>
>	Are you using /etc/shutdown, or the shutdown entry from the
>'install' login?  If not, or if you are not waiting for it to complete, you
>are PRODUCING problems for it to fix.  

He probably was running shutdown.  There are at least a couple wierd
things shutdown doesn't take care of that cause fsck to reboot.  I've
seen /etc/utmp come up with the wrong number of links and some core
files cause fsck a problem (I'm not sure the core files cause a
reboot though, they may just show up as DUPS).

>Just shuting down a unix computer is
>about the worst thing you can do for file system sanity.  (Except in some
>strange beasties, such as the Textronix 6130, and maybe its brethren, which
>have what is called a 'soft power switch'.)  When you turn it off, the led in
>it changes from steady to flashing, and the system starts a shutdown
>sequence.  It runs fsck before the shutdown completes, and sets a file as a
>flag that the fsck has been done.  Once this is all complete, it then pulls
>the rug out from under itself, and the led on the switch finally goes out.
>Boot time is reduced, since it already has a certified filesystem.  (Of

Is it fsck that is setting the flag, or umount?  Normally Unix has a
flag in the volume superblock that is set when a file is unmounted and
cleared when it is mounted.  Fsstat is run to check the status of,
among other things, that flag.  Depending on the status of the file
system, fsck is run if it appears that a normal unmount did not happen,
or not run if the flag is set.

Of course the root file system doesn't get unmounted, so fsck is always
run on it (a good idea anyway).

And on a UnixPc, which was designed to only have a root file system,
there is no flag, no fsstat, and fsck is always executed.

I'd guess most everyone with more than one hard disk (and multiple
file systems) has come up with a way to do the same thing by using
a file as a flag.  Lenny posted scripts to do this a long time back,
and I think they are on osu-cis.  I did mine a little different, but
the result is the same.  If a normal shutdown was done then only the
root file system is checked, but if it crashed fsck is run on every-
thing.

(I've learned via email that the floppy drive problem in the
original article did in fact have a "dust on the led" problem,
and does work now.  But his Teac 55BR drive still doesn't.)

Floyd
-- 
Floyd L. Davidson  |  floyd@ims.alaska.edu   |  Alascom, Inc. pays me
Salcha, AK 99714   |    Univ. of Alaska      |  but not for opinions.

zaft@suned1.nswses.navy.mil (Gordon C Zaft) (03/06/91)

	Hi!  Like Floyd reported, my OLD Teac 55B is now working fine, thanks
to a little judicious cleaning of the write protect sensor (not bad, since
I bought the machine with the assumption that it was bad (I'd been told it
didn't work)).

	The 55BR still doesn't work.  I'm assuming it's some kind of jumper
problem, since it just plain doesn't have as many jumpers on it as the
older drive.  I'm not too worried, though, since I have an older Teac
that has a marginal drive door that I could use as a desperation spare,
and the new 55BR could replace that one (it's currently in my PC).

	The multiple boot problem is still with me, but I had some 
ideas given me via e-mail as to things I could do.  I was indeed using
the shutdown from the menu and waiting until it said to turn the machine
off.

	Thanks a lot!  I got FIXDISK 2.0 put on the machine last night.
Next stop is HDB3.5 and setting myself up as a uucp site....


--
+  Gordon Zaft                        |  zaft@suned1.nswses.navy.mil         +
+  NSWSES, Code 4Y33                  |  suned1!zaft@elroy.jpl.nasa.gov      +
+  Port Hueneme, CA 93043-5007        |  Phone: (805) 982-0684 FAX: 982-8768 +
** Ray Kaplan is a God! *** "Quoth the raven: Eat My Shorts!" -- B. Simpson **