|
Grex > Helpers > #140: Grex System Problems - Spring 2005 | |
|
| Author |
Message |
| 25 new of 457 responses total. |
drew
|
|
response 256 of 457:
| 
May 4 21:01 UTC 2005 |
FWIW, I've had a disk *image file* (created with 'dd if=/dev/hdc of=filename')
produce read errors when used in the virtual machine it was attached to.
|
keesan
|
|
response 257 of 457:
|
May 4 21:14 UTC 2005 |
Three times now, with two different modems, we have dialed into grex and got
garbage. The second dial logged us in. Another grexer reports that the modem
on 484-0513 works but the first one does not, from his location. Is there
any other reliable modem that could be switched with the 0512?
|
steve
|
|
response 258 of 457:
|
May 4 23:03 UTC 2005 |
I think first we need to verify that the line and connection is OK,
physically. Sindi, do you know when these problems started? That
would be good to know.
|
cross
|
|
response 259 of 457:
|
May 5 00:53 UTC 2005 |
This response has been erased.
|
keesan
|
|
response 260 of 457:
|
May 5 01:02 UTC 2005 |
The garbage on dialin happened this week, probably in the last three days.
Jim mentioned it to me yesterday but I had already noticed. It might just have
started yesterday. It occurred again this afternoon.
Jim tried switching from 38 to 19K which did not help.
|
steve
|
|
response 261 of 457:
|
May 5 01:30 UTC 2005 |
Is it always the same modem that messes up?
|
albaugh
|
|
response 262 of 457:
|
May 5 15:22 UTC 2005 |
Drift: Does anyone else think that the fsck program name was partially chosen
because it looks like a get-past-the-censors-disguise for the f-word? ;-)
|
keesan
|
|
response 263 of 457:
|
May 5 16:11 UTC 2005 |
We always dial 0512 but I don't know which modem we actually reach. SOmeone
said the 0512 modem does not work for him but 0513 does, something about
distance from the phone company.
|
twenex
|
|
response 264 of 457:
|
May 5 16:42 UTC 2005 |
Re: #259 - ah, I see.
Re: #262 - Heh. I bet that is exactly the reason! :-)
|
tsty
|
|
response 265 of 457:
|
May 5 16:52 UTC 2005 |
hullo disk problems! there *IS* the best disk repair/recover
software for everyone - and now spinrite 6 will also do xnix drives
and mac drives.
the procedure for xnix formatted drives is a tad more detailed but
if you put *any* worth on your drives you simply must (sorry if that's
preachy) run spinrite on them about every 6 months.
i feel as if i *should* be preachign to the choir when i state adn
restate the obvious, but the choir is still out of tune, it seems.
spinrite 6 grc.com ............................... please!
|
tsty
|
|
response 266 of 457:
|
May 5 16:53 UTC 2005 |
oh, i also wnat to thankx STeve & company for all the extra efforts
on behalf of grex. thank you very mulch.
|
jor
|
|
response 267 of 457:
|
May 5 17:01 UTC 2005 |
Nonsense. None of that thank you stuff.
All we do is criticise, stamp our little feet,
and declare oursleves to be "paying members".
tsty get with the program dude.
|
steve
|
|
response 268 of 457:
|
May 5 18:12 UTC 2005 |
Heh...
The problem with disk "fixing" software is that is nearly all cases its
a giant kludge. With bit densities being hundreds of millions+ per square
inch, the most minute impurities left inside the disk case can cause disasters,
and the tolerances for everything mechanical has shrunk to amazing porportions.
This means that when something in a disk goes wrong its far harder to fix.
When bad disks come in the disk oem's look at the control electronics and
the disk case (mechanical) for problems. Depending on which they find bad
they throw that away and put another "known good" component in, test it and
then have a refurb disk for replacements. I'm not really happy with that
but thats the way things are.
Trying to alter a disks surface by rewriting something just isn't a good
idea now. Back in the era of 300M disks it worked to some extent.
Lastly, when you think of the sheer amount of data that you can put on
a 100G+ disk, you have a huge investment in that data, be it personal or
professional. It just doesn't make sense to trust kludges. Disks are too
cheap not to replace; data is too expensive to replace.
|
keesan
|
|
response 269 of 457:
|
May 5 18:57 UTC 2005 |
Regarding modems, I got garbage again dialing 0512 and before the garbage
there was briefly something about tty00. A second dialin immediately after
connected me properly - does this imply that I got the second modem this time
because the first was still tied up? If so can someone replace the first
modem, or at least confirm the problem? We got it at two locations.
|
cross
|
|
response 270 of 457:
|
May 5 19:31 UTC 2005 |
This response has been erased.
|
naftee
|
|
response 271 of 457:
|
May 5 19:41 UTC 2005 |
thanks for the mulch, tsty !
|
twenex
|
|
response 272 of 457:
|
May 5 19:44 UTC 2005 |
Snicker.
|
drew
|
|
response 273 of 457:
|
May 5 20:54 UTC 2005 |
Re #268:
You have a point about attempting to 'fix' a bad disk. However, a program
that gets the disk electronics to cough up the truth about how much of the
disk is *really* damaged, and how many reserve sectors are left, might be
useful for monitoring purposes.
Re #270:
Humor or not, it may not be far from the truth. cf. HTML content, flash
animations, spam, etc.
|
steve
|
|
response 274 of 457:
|
May 6 05:41 UTC 2005 |
Drew, the problem with that is when things are damaged, how can you trust
the electronics? As an example, IBM has something called smart for their
disks. It's a system where you can run a drive fitness test on a disk to
get a sense of its health. I've found it to be useful in telling me whats
wrong with a dead disk, usually. But it has failed me several times when
testing a disk that the user said had acted weirdly. To be fair, it did
catch a disk that was on the verge of going bad, but I still think the
technology is ripe for improvement.
|
gull
|
|
response 275 of 457:
|
May 6 15:00 UTC 2005 |
Re resp:270: It's definitely true where I work.
I'm not sure how valuable Spinrite-style products really are these days.
You can't directly address sectors on a disk anymore for testing -- the
drive electronics hide all those details and remap bad sectors from a
pool of spares. By the time there are actually visible bad sectors, the
disk has been going south for a long time.
|
tsty
|
|
response 276 of 457:
|
May 6 16:18 UTC 2005 |
re #267 ... oh, right, i forgot. stomp sTomP SToMp st0MP, ds al coda
ummmmmmmmm, about spinrite. every 'objection/dismissal' above demonstrates
that not one of you has read up on *what* it does nor *how*! dammit!!
there is *NO* comparable program in the universe. it FSCKING works!
and all the hidden shit is bypassed, obviated, shunted, circumvented,
counteracted and evaded <insert further descriptions here>.
steve gibson is long overdue for a macarthur grant, imnsho.
and one of it's best built-ins is that it catches stuff and fixes it
BEFORE shit hits the fan. /sheeeeeeeeeeeeeeeeeeeeeeeeeeeSH!
|
russ
|
|
response 277 of 457:
|
May 8 00:13 UTC 2005 |
Configuration of the new disk isn't quite done; /var/log/wrttmp isn't there.
|
richard
|
|
response 278 of 457:
|
May 8 01:33 UTC 2005 |
I have noticed that now when you !finger anybody to see when they logged in
last, it says "never logged in" For any user that I've tried. I guess all
recent login information has been lost?
|
keesan
|
|
response 279 of 457:
|
May 8 01:57 UTC 2005 |
I got garbage again the first time I dialed in but different looking garbage,
and the same wrttmp message.
|
drew
|
|
response 280 of 457:
|
May 8 03:30 UTC 2005 |
Re #274 and 275:
Out of curiosity I checked out a copy of Spinrite. It seems to do the
usual sector checking and attempted 'fixing', with five different levels of
intervention. But in addition, I've found a screen called "SMART settings"
that went something like this:
attribute event cnt margin
---------------------------------
ecc corrected: 0 149
rd chan margin: not reported
relocated sect: 0 60
realloc events: 0
seek errors: 0 49
spin-up retry: 0 49
recal retries: 0 49
cabling errors: 0
uncorrectable: 0
write errors: 0 149
temperature: 40'c /104'f
power-on time: 5,271
Not sure it means anything; a 'margin' of 149 sectors out of 63*255*thousands
seems rather tiny. But I think this is supposed to be the 'bypassing the drive
electronics' part of the program.
Spinrite also works on VMWare virtual machines. However, the SMART screen
does not appear; instead there's a message saying that SMART data is not being
reported.
|