Re: Single-bit corrected errors



On Mon, 27 Feb 2006 18:48:50 -0600, willbill wrote:

Keith wrote:

I've been moving disk drives (DVD and HDD) around and upgrading OS
(SuSE 9.0 to 10.0) over the past couple of days.

i've got the same Tyan AMD/940 mobo as you (S2875s), and
i remember your fairly recent use of 9.0 (and problems
of some sort?)

I got 9.0 to be pretty stable. There were things I didn't like about it
(USB sticks were a PITA), but never had any real issues. The ECC
problems have been under the covers since day one, pretty much. I only
looked in the BIOS for hints trying to track down other problems. The
error los were full and dated back at least a year, perhaps to day-one.

what exactly prompts you to go to 10.0?

USB sticks, primarily. Possible SATA support was another reason. A
workign DVD player? Nah!

USB sticks now work, at the cost of dual-display. :-( ...and DVDs still
might just as well be blank. I haven't gotten around to test SATA yet.
After a couple of years, the drive is still just heating the room.

i ass-u-me you think it's gonna solve yer 9.0 problem(s)?

USB sticks work. Dual display broke. :-( SATA is still a maybe. Again,
9.0 was pretty stable (ok, so I have to get my head examined...).

i mean, did you buy 10.0 at full-retail/discount, or from a
discount/show/friend/e-bay source or what?

Full retail boat: $59.95 plus 7% sales tax. :-( ...from Best Buy, no
less! <shudder>

as you might guess, i've got some interest in going back into Linux
(i've used Unix on/off since '89)

I sorta like it for some things, not so for others. My Windows machine
has been sitting in the corner for a couple of months waiting for time to
get all the frappin' cables set up, if that gives you a clue how much I
*need* WinBlows. OTOH, SHMBO kinda wants to get a (film capable) scanner
digital cameras, and a *PRINTER* (her emphasis) working, so...

and yes, i'm posting this via win98se <big grin!> and Mozilla 1.7.7
(i've been on the bleeding s/w edge too many times!; i've learned my
lesson! :) )

SuSE 10.0 isn't exactly bleeding-edge. It's been out for a while. I
didn't go to 10._later_than_zero because it _is_ alpha or beta.

In the process
I've been playing with BIOS to get the drives ordered the way I wanted
(and figure out who listens to BIOS drive ordering ;). While trying to
figure out some strangeness I cleaned out the BIOS messages area. Last
night I kept getting beeps (about one an hour). It took a while to
isolate it, but finally looked at the hardware error report in the
setup screens. Seems I've been getting a corrected single-bit error in
the same location on average about once an hour. I guess ECC actually
does something. ;-)


sorry that it's happened to you, but it's good to hear that it really
works, from a reliable source, namely: you

Aw <blush>. Actually, I'm kinda interested in this stuff now. Before it
was a matter of "it works or it doesn't". ...only her hairdresser knows
for sure. Now the hairdresser has come out of the closet and we
still don't know what s/he is. ;-)

like you, i suspected that it really worked, but like you, i had my dark
doubts

Exactly. Now I'm shocked (*shocked*, I tell ya') that they can't pinpoint
the problem, given the error code. OTOH, I'm sorta pleased that Tyan has
given me the time of day; answers, such as they are, over the weekend
even.

....now to figure out which stick is faulty (it's all Crucial, with a
lifetime warranty ;-). If everything is right in the world the error
report _should_ point me to it. All I need is the Rosetta stone.


can't help you with which stick

You're not alone. ;-)

especially can't help you with Rosetta stone

<"A basalt tablet bearing inscriptions in Greek
and in Egyptian hieroglyphic and demotic scripts that was discovered
in 1799 near Rosetta, a town of northern Egypt in the Nile River
delta, and provided the key to the decipherment of Egyptian
hieroglyphics.">

i grant that PC computers (whichever OS) are like hieroglyphics. :)

LOL! Sometimes I think people purposely treat what they know as such a
secret (one programmer refused make his source "public" - didn't last
long) to protect their jobs. Too bad, because the ones who know what's
going on needn't protect anything. Those who hide, have a reason and
aren't needed. Has noone heard the old programmers saw; if you don't know
what to do with an error, don't test for it? ;-)

--
Keith
.



Relevant Pages

  • Re: Boot from USB PRM (Partitioned Removable Media ) failed after loading OS2LVM.DMD error: "OS/2 i
    ... I've never even tried the USB external hard drive ... The drivers involved are the same. ... OS2DASD.DMD usually serves the fixed drives first, ... BIOS which shifts the PRM before the fixed disks. ...
    (comp.os.os2.bugs)
  • Re: Master/Slave (S & M?)
    ... > I want a way to occasionally boot from the new slave HD without opening ... Can I do this thru the BIOS? ... low level, DOS like support, for USB keyboards and mice. ... drives. ...
    (microsoft.public.windowsxp.basics)
  • Re: Booting up question
    ... drives, CD/DVD, floppy and so forth. ... Sequence of double bios execution ... Unless you're using some intermittent option to boot from the USB ...
    (microsoft.public.windowsxp.general)
  • Re: Booting up question
    ... drives, CD/DVD, floppy and so forth. ... Sequence of double bios execution ... Unless you're using some intermittent option to boot from the USB ...
    (microsoft.public.windowsxp.general)
  • Problems with HP dx5150/ATI Xpress 200 chipset
    ... ATI USB ... For all the installs I made the following changes to the default BIOS ... basic functionality, including network working. ... APIC settings did not seem to affect obsd boot or installs, ...
    (comp.unix.bsd.openbsd.misc)