Dumb like a fox Backup Software?
Frank Stearns
save-set formats, but I learned years ago never to fully trust those.
Migrate computers, or lose the original software (or have it not run on your fancy
new hardware), and you're screwed, or at least in for a tedious recovery process.
Plus, it's often hard to pluck out the one or two files you might quickly need
because you can't readily see/search for them without first "decoding" the save-set.
Also, any third party receiving these disks would need the same backup software that
you might have used to read them.
What I'd love is backup software that will simply write a directory structure onto
the media of choice, each member file in its native format, with enough wits about
it to prompt for the next disk as it goes, retaining the original directory
structure as it goes.
Double points if it does a pre-scan of the structure you wish to backup, then on the
first disk write a plain text file listing what it plans to put on what disk in the
back up set of disks.
Triple points if it's smart about media problems and doesn't make you start over
again should one of the backup disks have some sort of problem.
I know this isn't a terribly efficient method, but I'll trade that for (a) not
needing "decode/recovery" software and (b) the ability to immediately access the
files in their original format.
Anyone know if such a beast exists?
Thanks in advance,
Frank
Mobile Audio
--
.
John Williamson
two directories synchronised. It doesn't compress or do anything clever,
just uses the standard Windows file copying routines.
http://www.brothersoft.com/synchronx-4732.html
It compares all file modification dates in any two directories, and
copies the latest version and all new files to the other directory. It
will also scan and mirror sub-directories if required. It can be used to
transfer the entire contents of a directory to a new machine. It also
works across a network.
It's also free, and under half a megabyte download.
The *only* problem I find is that where you have a FAT HD and an NTFS
HD, the dates are not respected, as there's about a week's discrepancy
in the reported modification dates. There's probably a work-round, other
than converting the FAT HD to NTFS, but I've not found it yet.
--
Tciao for Now!
JOhn.
Mike Rivers
> I know this isn't a terribly efficient method, but I'll trade that for (a) not
> needing "decode/recovery" software and (b) the ability to immediately access the
> files in their original format.
>
> Anyone know if such a beast exists?
I've been using Norton Ghost 2003 (bought an old copy for $15 on a
closeout) and
haven't had to use it in any emergencies. It makes (in the mode I use
it) a full clone
of the disk you're copying with it. I've successfully used it when
upgrading
the disk drive in a computer by making a clone of the existing drive and
swapping
it out with the clone.
You have to be careful when using a clone to migrate to a new computer,
though. It
will have the hardware drivers installed for the hardware on the old
computer, and
that may not be correct for the new computer. Things like the graphics
display or sound
card, and on occasion CD drive won't work right until you un-install the
device (I'm
talking Windows here), restart the computer, and let it find the
hardware and look for
the proper driver. You may need to feed it the Windows installation CD
if there isn't
a copy on the hard drive.
Laurence Payne
wrote:
>You have to be careful when using a clone to migrate to a new computer,
>though. It
>will have the hardware drivers installed for the hardware on the old
>computer, and
>that may not be correct for the new computer. Things like the graphics
>display or sound
>card, and on occasion CD drive won't work right until you un-install the
>device (I'm
>talking Windows here), restart the computer, and let it find the
>hardware and look for
>the proper driver. You may need to feed it the Windows installation CD
>if there isn't
>a copy on the hard drive.
In the worst case the computer won't boot at all. The Registry goes
looking for hardware that isn't there, doesn't find it and baulks. The
problem is likely to be motherboard components rather than add-ons
like graphics cards and drives - Windows knows how to cope with
changes of this sort.
If this happens there are ways of deleting parts of the Registry so
the different hardware can be detected and installed.
You're quite likely going to need the drivers disk that came with the
motherboard to get things like onboard network ports and audio
working. Or download them from the board maker's website, where they
are inevitably available. Quite a good idea to do this BEFORE
considering a computer change :-)
Frank Stearns
I perhaps should have given a bit more background. I own Norton Ghost, have used
zip, played with other 3rd party apps that do "backup" and can span optical media,
and have even used scripts based on Tar, UU, and Split back in the Sun-OS UNIX days.
But that's not what I'm looking for in this case. I'm not moving to a new machine,
just producing project datasets for clients. (My point about machine migration is
that when you do so you then run the risk of creating orphan datasets should they be
in some special format, because the original encoding scheme might no longer be
supported. And then you *still* can't see invidual files without "opening" the
dataset.)
My data target is mainly 4.7 Gbyte DVD-Rs, primarily as a way to offload a project
either to physical file archive (put a couple/three DVDs in with the paperwork in a
file folder), or hand off something relatively cheap and slightly more
take-and-forget robust than an HD to a client (assuming they don't leave the disks
in their car on a sunny day).
With most of what I do, client projects run between 8 and 20 Gbytes. With larger
projects I ask that the client purchase a porta HD of some kind, then warn them to
spin it up once or twice a year or risk losing the data.
Given that I've now read data from 14 year old optical media with no problems, I'm
hoping that this method will stand up to time slightly better than HDs.
These projects do not have the budget for my active storage management here (nor do
I want to be getting calendar reminders in 2016 to do yet another HD spin up/write
to new HD for the local choral society).
Nor can I get too esoteric. Small bundles of DVD-Rs with files in their native
format in the original folder structure, seems a good way to go. I've built such
datasets by hand, but it's a distraction.
Surely determining how to split a dataset and do the other items I mention is
just the kind of mundane thing well suited to a computer!!!
Here's hoping, anyway...
Thanks again; let me know if something else comes to mind.
Frank
Mobile Audio
>Frank Stearns wrote:
>> There are many flavors of backup software that will write compressed/proprietary
>> save-set formats, but I learned years ago never to fully trust those.
>>
>> Migrate computers, or lose the original software (or have it not run on your fancy
>> new hardware), and you're screwed, or at least in for a tedious recovery process.
>> Plus, it's often hard to pluck out the one or two files you might quickly need
>> because you can't readily see/search for them without first "decoding" the save-set.
>> Also, any third party receiving these disks would need the same backup software that
>> you might have used to read them.
>>
>> What I'd love is backup software that will simply write a directory structure onto
>> the media of choice, each member file in its native format, with enough wits about
>> it to prompt for the next disk as it goes, retaining the original directory
>> structure as it goes.
>>
>> Double points if it does a pre-scan of the structure you wish to backup, then on the
>> first disk write a plain text file listing what it plans to put on what disk in the
>> back up set of disks.
>>
>> Triple points if it's smart about media problems and doesn't make you start over
>> again should one of the backup disks have some sort of problem.
>>
>> I know this isn't a terribly efficient method, but I'll trade that for (a) not
>> needing "decode/recovery" software and (b) the ability to immediately access the
>> files in their original format.
>>
>> Anyone know if such a beast exists?
--
.
Mike Rivers
> just producing project datasets for clients.
> My data target is mainly 4.7 Gbyte DVD-Rs, primarily as a way to offload a project
> either to physical file archive (put a couple/three DVDs in with the paperwork in a
> file folder), or hand off something relatively cheap and slightly more
> take-and-forget robust than an HD to a client (assuming they don't leave the disks
> in their car on a sunny day).
As long as you're just copying data, there's nothing safer than just
plain Copy, to
as many disks as it takes. Don't use any special software.
> Surely determining how to split a dataset and do the other items I mention is
> just the kind of mundane thing well suited to a computer!!!
It's not that much trouble to split things up logically by eye. Put
related things in
folders if that makes sense for your data, sort the folder by file size,
and just start
copying from the top down. Don't worry if you might have been able to
squeeze one
or two more small files from the end of the list together with some
large files. It's
too confusing and doesn't save that much media space.
lowg...@ao1.com
No. From within a running Windows system, many files cannot be "copied"
because they are seriously in use.
Windows has special API calls to access ALL files. It is used by modern
backup software. I forget what the subsystem is called.
Richard Webb
> What I'd love is backup software that will simply write a directory
> structure onto
> the media of choice, each member file in its native format, with
> enough wits about
> it to prompt for the next disk as it goes, retaining the original
> directory structure as it goes.
I did something like this with modifications to a batch file I found in PC COmputing magazine back in 1990, originally to floppies, then to the 100 mb zip disks. IT used the dos
xcopy command in its original iteration, with command line
switches that set the archive bit on copied files, when the
media was full prompted you to insert another then continued where it left off because of the setting of the archive bit. ON the first disk, before we started copying files it
generated a text file of what was to be backed up, i.e. the
contents of the source directory.
This could be easily modified to use cd-r or dvd-r media I"d think, except using a target empty directory, and when
contents were of sufficient size to fill the media goes into burn your disk mode. I think later versions of windows have something like xcopy which verifies the correct writing of
the file to the destination.
IT should work even in xp etc. I'd think, because no
environment variables need be set. Just set the archive bit on the files copied.
<snip>
> Triple points if it's smart about media problems and doesn't make
> you start over again should one of the backup disks have some sort
> of problem.
THis was, it just said it couldn't write to that disk, zero
files copied and prompted for another disk.
I've got an older iteration of that program here, sitll
designed for floppies in drive a: but could be modified to
fit your usb hard drive or whatever media you choose.
IF interested, drop a note to elspider at bellsouth dot net
Frank and I'll shoot it at you via that one.
Regards,
Richard
--
| Remove .my.foot for email
| via Waldo's Place USA Fidonet<->Internet Gateway Site
| Standard disclaimer: The views of this user are strictly his own.
Frank Stearns
>Frank Stearns wrote:
>> just producing project datasets for clients.
>> My data target is mainly 4.7 Gbyte DVD-Rs, primarily as a way to offload a project
>> either to physical file archive (put a couple/three DVDs in with the paperwork in a
>> file folder), or hand off something relatively cheap and slightly more
>> take-and-forget robust than an HD to a client (assuming they don't leave the disks
>> in their car on a sunny day).
>As long as you're just copying data, there's nothing safer than just
>plain Copy, to
>as many disks as it takes. Don't use any special software.
Thanks, Mike -
But we're talking about loading stuff into DVD burner software, ala Nero or one of
those. (The operation is much like copy if you drag and drop from explorer to the
burn list.)
What you say is *exactly* what I do now by hand, but it is a distraction, and the
chance of inadvertantly leaving out a file as I do the "manual split" is one of the
items that bothers me.
Sure, it can all be done, and reliably, but it takes up a lot of productive time if
you have a slew of projects to archive, backup, or get to clients, and you want to
be absolutely sure the burn set has full integrity.
In my best Lt. Commander Scott voice, I want to (in essence) say: "Computer! Back Up
the Choral Arts Society November concert to DVD. Let me know when you want the next
blank disk."
These kinds of low-level tasks -- doing the data set split-up while retaining the
original directory stucture, tracking what got burned on what disk, prescanning the
save set to support writing a split list plain text file on disk #1, etc, etc -- are
exactly what our idiot silicon friends should be really good at, but durned if I can
find such an application! (One that doesn't compress or write a proprietary binary
format of the source files -- we already have lots of those.)
>It's not that much trouble to split things up logically by eye. Put
Except that the first time you get a phone call, or take on some other distraction,
you might misstep on the copy process. Sure, you can go back to the previous disk to
check, but once again we're burning daylight on something the machine should do.
>folders if that makes sense for your data, sort the folder by file size,
I want to retain the original folder structure, so that "restoration" means nothing
more than copying disk after disk to the same "root" directory for a recovered
project (usually the projects are in Protools, but there are also mastered versions
of the 2 tracks, with graphics and session documentation).
>folders if that makes sense for your data, sort the folder
>and just start
>copying from the top down. Don't worry if you might have been able to
>squeeze one
>or two more small files from the end of the list together with some
>large files. It's
>too confusing and doesn't save that much media space.
Agreed completely. I don't care about wasting some (or even a lot) of space on a 50
cent DVD if it means easy and robost visits to the file set.
Thanks again,
Frank
--
.
Mike Rivers
> No. From within a running Windows system, many files cannot be "copied"
> because they are seriously in use.
Why would you want to back up a file that was in use? This is data,
remember?
Quit whatever is using the files and then copy them.
Soundhaspriority
"Frank Stearns" <franks.pa...@pacifier.net> wrote in message
news:cPmdnUFtRpv8XonW...@posted.palinacquisition...
> There are many flavors of backup software that will write
> compressed/proprietary
> save-set formats, but I learned years ago never to fully trust those.
>
> structure onto
> the media of choice, each member file in its native format, with enough
> wits about
> it to prompt for the next disk as it goes, retaining the original
> directory
> structure as it goes.
>
About 20 years ago, there was a crossover in price, when storage on
removable media became more expensive than hard disk. Of late, the cost
differential has become so extreme, many people, perhaps the majority, have
abandoned the use of removable media.
What you want is going against the grain. With computers, there is
surprisingly high cost of doing this. Let's examine the chances of losing
data archived to a hard disk. If the hard disk has a 5% chance of failing in
one year, that's .013% chance of this happening per day. If you have
duplicated the data onto two disks, the chances of failure simultaneous to
the day are one in 10 to the minus 8.
DVDs, with exposed surfaces, unstable dyes, and degradation over time
with or without high humidity, are not paragons of virtue.
I found it initially difficult to give up the concept of write-once
storage to a media I could hold in my hand. I did not want to lose the
mental association between data, and a physical object. But the world has
gone this way, for very logical reasons.
Bob Morein
(310) 237-6511
Mike Rivers
> What you say is *exactly* what I do now by hand, but it is a distraction, and the
> chance of inadvertantly leaving out a file as I do the "manual split" is one of the
> items that bothers me.
Hire an intern. Do it at the end of the day. Don't let the job pile up
(unless you have
an intern). I don't like to let computers take charge of too much,
particularly when it's
something I'm delivering to a customer. If you're really that busy, you
NEED and intern
or an assistant.
> These kinds of low-level tasks -- doing the data set split-up while retaining the
> original directory stucture, tracking what got burned on what disk, prescanning the
> save set to support writing a split list plain text file on disk #1, etc, etc -- are
> exactly what our idiot silicon friends should be really good at, but durned if I can
> find such an application!
How would you tell a computer how to do that? If you were doing it by
hand, you'd
be making some decisions:
If I can fit two directories worth of files on to one disk, I'll do
that. So what if it's only 75% full.
If I can't even fit one directory on a disk, then I'll have to split it.
Might as well take the big files first
If I have a bunch of little directories, they can go on a disk.
I don't want to put any "stray" files on any disk
I wouldn't know how to program that.
> I want to retain the original folder structure, so that "restoration" means nothing
> more than copying disk after disk to the same "root" directory for a recovered
> project (usually the projects are in Protools, but there are also mastered versions
> of the 2 tracks, with graphics and session documentation).
Sounds like you need a disk for the "mastered and documented" version
and as many
other disks as it takes to keep the directories intact. Hopefully no
directory is larger than
one DVD or whatever you're using as your backup.
I just don't see it as being that hard. Do it when you're cleaning up so
you can do something
mindless between dragging files to the burning program. Or just get a
$50 USB external
hard drive, put everything on it, and let the customer sort it out when
he needs it. I'm doing
that more and more these days for my own backups.
Frank Stearns
>"Frank Stearns" <franks.pa...@pacifier.net> wrote in message
>news:cPmdnUFtRpv8XonW...@posted.palinacquisition...
>> There are many flavors of backup software that will write
>> compressed/proprietary
>> save-set formats, but I learned years ago never to fully trust those.
>>
>> What I'd love is backup software that will simply write a directory
>> structure onto
>> the media of choice, each member file in its native format, with enough
>> wits about
>> it to prompt for the next disk as it goes, retaining the original
>> directory
>> structure as it goes.
>>
>Frank,
> About 20 years ago, there was a crossover in price, when storage on
>removable media became more expensive than hard disk. Of late, the cost
>differential has become so extreme, many people, perhaps the majority, have
>abandoned the use of removable media.
> What you want is going against the grain. With computers, there is
>surprisingly high cost of doing this. Let's examine the chances of losing
>data archived to a hard disk. If the hard disk has a 5% chance of failing in
>one year, that's .013% chance of this happening per day. If you have
>duplicated the data onto two disks, the chances of failure simultaneous to
>the day are one in 10 to the minus 8.
This *assumes* you spin the HD at least a couple of times per year. After 12-18
months of not having been spun, we run the risk of the mechanics failing.
Then there's the cost for 8-20 Gbytes of storage. Yes, HD storage IS cheap, but if
you only use a few percent of the drive capacity, your cost per unit of used storage
goes up.
And then, I can't easily slip a few DVDs into the file folder with the session
docs!! :) I suppose you could do that with a 2.5" drive, but then there are
mechanical risks with that as well.
I'll still have HD archival of ALL client material somewhere, but I have been
impressed that optical media *can* last (at least longer than an un-spun HD), per
my comment about 14 year old burned media (CDrs, both 74 and 80 minute flavors).
So when those HDs are forgotten and frozen, say 10 years from now, a set of DVD-rs
might still be useable.
> DVDs, with exposed surfaces, unstable dyes, and degradation over time
>with or without high humidity, are not paragons of virtue.
Agreed, but if quality -Rs are used (NOT the RW variety/off-brand cheapie), stored
with a little care in the padded/cleaner sleeves I use, the storage life might be
quite a bit longer.
NONE of these solutions is really all that great; I'm just trying to find the right
compromise until such time as we have robust, 50-100 year media, as we do/did with
analog tape, pre-456 formulations. And no, my clients aren't going to pay for a
transfer to analog tape!! :)
> I found it initially difficult to give up the concept of write-once >storage to a
media I could hold in my hand. I did not want to lose the >mental association
between data, and a physical object. But the world has >gone this way, for very
logical reasons.
Not quite sure I follow... Indeed, the world has gone certain ways for certain
reasons (and I do use those solutions for a portion of my work), but it's not a one
size fits all, IMO.
Looks like I might have to cobble my own solution, based on Richard's script.
Thanks again for the replies,
Frank
--
.
Frank Stearns
>Frank Stearns wrote:
>> What you say is *exactly* what I do now by hand, but it is a distraction, and the
>> chance of inadvertantly leaving out a file as I do the "manual split" is one of the
>> items that bothers me.
>Hire an intern. Do it at the end of the day. Don't let the job pile up
>(unless you have
>an intern). I don't like to let computers take charge of too much,
>particularly when it's
>something I'm delivering to a customer. If you're really that busy, you
>NEED and intern
>or an assistant.
Mike!? Now we're serving the machine!! I have to hire somebody to do what the
machine should be able to do with ease???
Maybe it's just me, but I get the same queasy feeling from that as when I was
punching Fortran decks in 1971! <g>
>> These kinds of low-level tasks -- doing the data set split-up while retaining the
>> original directory stucture, tracking what got burned on what disk, prescanning the
>> save set to support writing a split list plain text file on disk #1, etc, etc -- are
>> exactly what our idiot silicon friends should be really good at, but durned if I can
>> find such an application!
>How would you tell a computer how to do that? If you were doing it by
The logic is simple. Start interating the target file structure in a scanning pass.
Add up file sizes as you go. When you hit the magic number (4.7 Gbyte less
anticipated overhead less whatever other margin you wanted), time for a new disk.
Didn't finish the current directory? No problem! Create that part of the tree again
on the next disk, and continue with files from that folder that didn't make it on
the last disk.
Keep doing that until done. Generate the summary plain text file. Include a review
step if you wish (I like apps that do that kind of thing). Go back and do the
burning "for real".
>I wouldn't know how to program that.
It's not overly hard, just another distraction to sit down and code the thing. <g>
>> I want to retain the original folder structure, so that "restoration" means nothing
>> more than copying disk after disk to the same "root" directory for a recovered
>> project (usually the projects are in Protools, but there are also mastered versions
>> of the 2 tracks, with graphics and session documentation).
>Sounds like you need a disk for the "mastered and documented" version
>and as many
>other disks as it takes to keep the directories intact. Hopefully no
>directory is larger than
>one DVD or whatever you're using as your backup.
So long as we keep maintaining the directory structure disk-to-disk, it doesn't
matter which files are where, because during any "restore" process we'll keep
copying each disk back to the same target "root" folder, each disk making its
contribution to the tree. (Think of it as a static cousin to the NFS file system, if
you're familiar with the old NFS approach used by Sun Microsystems.)
The ONLY hassle would be a *file* larger than 4.7 Gbyte, and best I know no such
critters exist. We have BWV files, but aren't they simply chained-together 2 GByte
files?
>I just don't see it as being that hard. Do it when you're cleaning up so
It's not hard, just annoying when we have these computational wonders that no
one has (apparently) told how to sum file sizes and check whether the next addition
will make the total greater than some number.... It's actually rather amusing in a
way...
>you can do something
>mindless between dragging files to the burning program. Or just get a
>$50 USB external
>hard drive, put everything on it, and let the customer sort it out when
>he needs it. I'm doing
>that more and more these days for my own backups.
Been doing that here too for a long time; I have an entire library of backup drives
and three machines with multiple Tbytes online to "round robin"
active/archival/backup data and spread the risk.
And the larger clients *do* have to get their own porta drives.
This is a special application with a specific purpose, targeting a certain file set
size.
RD Jones
wrote:
Second Copy does most of what you are asking.
rd
Les Cargill
I wrote my own. If/when I actually get a Linux box up,
I'll just use rsync. rsync for windows.... well, not
really...
Thinking again, one could use rsync from a Knoppix
boot disk.
--
Les Cargill
Mike Rivers
> This *assumes* you spin the HD at least a couple of times per year. After 12-18
> months of not having been spun, we run the risk of the mechanics failing.
I remember the old ST-251 disks used to be like that, but are modern ones
still like that? If so, I probably have a few "reels of tape" from my
hard disk
recorders that need to be exercised.
> Then there's the cost for 8-20 Gbytes of storage. Yes, HD storage IS cheap, but if
> you only use a few percent of the drive capacity, your cost per unit of used storage
> goes up.
Actually, today the problem is FINDING hard drives that small. But
still, the cost
for a drive is pretty darn small. Fifty bucks will buy you enough space
for one,
or several projects.
> And then, I can't easily slip a few DVDs into the file folder with the session
> docs!! :) I suppose you could do that with a 2.5" drive, but then there are
> mechanical risks with that as well.
So you put it in a box. When Mackie came out with the HDR24/96, they had a
package for the removable drive (a standard 3-1/2" hard drive in a
mobile rack)
that was an anti-static foam lined case the same size as a U-Matic
videocassette.
There was room for several sheets of paper, you could slip a cover sheet
behind
a protector on the outside of the case with the essential identification
data on
the front, back, and spine, and it would fit on a shelf like a book. And
before you ask,
I've never found those cases anywhere else but from Mackie. I tried using a
U-Matic case, but it has bumps molded into the back to block the reels.
Fooey!
> I'll still have HD archival of ALL client material somewhere, but I have been
> impressed that optical media *can* last (at least longer than an un-spun HD), per
> my comment about 14 year old burned media (CDrs, both 74 and 80 minute flavors).
No argument there. For a while I was using double DVD cases to hold two
CDs for
a typical 24-track project's worth of data.
> So when those HDs are forgotten and frozen, say 10 years from now, a set of DVD-rs
> might still be useable.
I have analog tape that's more than 50 years old that still plays. Pffffft!
Mike Rivers
> It's not hard, just annoying when we have these computational wonders that no
> one has (apparently) told how to sum file sizes and check whether the next addition
> will make the total greater than some number.... It's actually rather amusing in a
> way...
Sometimes I think operating systems have become dumber. When I want to
print out
a listing of files on a directory, in Windows, I exit to a DOS shell and
type
dir d:\hdrprojects\fireflies\audiofiles\*.* >prn
I don't know how to do that in Windows. Never did. Probably never will.
Probably you can't.
Soundhaspriority
"Mike Rivers" <mri...@d-and-d.com> wrote in message
news:hf4p2u$6ht$1...@news.eternal-september.org...
> Frank Stearns wrote:
>
>> This *assumes* you spin the HD at least a couple of times per year. After
>> 12-18 months of not having been spun, we run the risk of the mechanics
>> failing.
>
> I remember the old ST-251 disks used to be like that, but are modern ones
> still like that?
silicone. If the head sat in one spot long enough, "sticktion" would occur.
At many companies in the 80's, the turn-on ritual for a Sun "pizza box"
involved dropping it onto the desk from a height of three inches, to break
the sticktion.
Modern hard disk platters have thin film substrates, coated with evaporated
diamond. There is no lubricant, and nothing to cause stiction. But when
turned off, the heads don't even sit on the disk. In Seagate drives, the
heads sit on a stippled landing zone surface, textured by a laser. In
Western Digital disks, the heads sit on low friction plastic loading ramp.
Bob Morein
(310) 237-6511
Frank Stearns
>Frank Stearns wrote:
Well, you just did! (a "DOS Shell" is now a "command" window and that's exactly the
syntax you'd use.) Add in true regular expression parsing and you'd darn near have a
real OS. (The DOS command line interface does some bozo stuff, and has a small
fraction of the power of true regular expressions. Heck, I could easily write my
little backup program in a shell or awk script.)
OSs are indeed in many ways far less useful then they used to be, so as to make
things "easier"... read as: "don't make me think or learn anything, just let me
click on pretty pictures". That's fine for a while, and even works for several
things, but gets bloody tedious if you have a lot of things to do (such as your
printed listing example).
That's what I loved so much about my days with SunOS (UNIX) or even VAX/VMS. You
took about a half-dozen or so primitive commands and you could do some very useful
stuff by stringing them together in different ways.
The best of bygone OS days was the NeXTStep OS -- A useful, mostly reliable GUI
sitting on top of Berkeley (or was it AT&T?) UNIX, where you could move between
graphical and command line uses with ease. SunOS wasn't half bad either.
But I digress...
Frank
--
.
Frank Stearns
This could be good news... So there are no lubricants of any kind anywhere in the
drive that could congeal/get tacky/whatever if not "used"?
Are you saying that the need for spin up is "no more", and the drive could sit on
the shelf for, say, a decade and spin up and be read okay?
If true, that would be a relief...
Frank
--
.
Soundhaspriority
viscous layer. The surface is pure diamond.
Bob Morein
Frank Stearns
- self snips -
>On Dec 1, 2:18=A0am, Frank Stearns <franks.pacifier....@pacifier.net>
>>
>> What I'd love is backup software that will simply write a directory struc=
>ture onto
>> the media of choice, each member file in its native format, with enough w=
>its about
>> it to prompt for the next disk as it goes, retaining the original directo=
>ry
>> structure as it goes.
- more snips -
>Second Copy does most of what you are asking.
Thanks, RD!
This does look quite promising for what I have in mind.
I'll be checking this out more thoroughly soon.
Thank you again,
Frank
--
.
Laurence Payne
wrote:
>Sometimes I think operating systems have become dumber. When I want to
>print out
>a listing of files on a directory, in Windows, I exit to a DOS shell and
>type
>
>dir d:\hdrprojects\fireflies\audiofiles\*.* >prn
>
>I don't know how to do that in Windows. Never did. Probably never will.
>Probably you can't.
You've just described how you did it! It's not really "exiting", that
sounds as if you're re-booting into DOS! The Command Prompt runs like
any other Windows application. You don't even need to type a path -
drag a folder onto the Command Prompt window and see what happens.
Arny Krueger
> There are many flavors of backup software that will write
> compressed/proprietary save-set formats, but I learned
> years ago never to fully trust those.
>
> Migrate computers, or lose the original software (or have
> it not run on your fancy new hardware), and you're
> screwed, or at least in for a tedious recovery process.
> Plus, it's often hard to pluck out the one or two files
> you might quickly need because you can't readily
> see/search for them without first "decoding" the
> save-set. Also, any third party receiving these disks
> would need the same backup software that you might have
> used to read them.
>
> What I'd love is backup software that will simply write a
> directory structure onto the media of choice, each member
> file in its native format, with enough wits about it to
> prompt for the next disk as it goes, retaining the
> original directory structure as it goes.
In these days of relatively inexexpensive hard drives, it is both possible
and reasonable to simply clone a hard drive in a few hours.
The clone of course has the same files and file structure as the original,
and is itself bootable with all software and hardware drivers installed.
> Double points if it does a pre-scan of the structure you
> wish to backup, then on the first disk write a plain text
> file listing what it plans to put on what disk in the
> back up set of disks.
Your documentation of the original is any report you choose to run on the
clone.
> Triple points if it's smart about media problems and
> doesn't make you start over again should one of the
> backup disks have some sort of problem.
IME hard drives are pretty reliable, and many many generations of backup can
be performed on them with perfect reliability.
Another more dynamic approach is to do disk mirroring. Each disk is a
perfect clone of the other.
Mike Rivers
>> dir d:\hdrprojects\fireflies\audiofiles\*.* >prn
>> I don't know how to do that in Windows. Never did. Probably never will.
>> Probably you can't.
>
> Well, you just did! (a "DOS Shell" is now a "command" window and that's exactly the
> syntax you'd use.) Add in true regular expression parsing and you'd darn near have a
> real OS.
I just figured that when using the Windows (file) Explorer there would
be a drop-down
menu selection to print what you see, like most every other Windows
program.
Mike Rivers
> The Command Prompt runs like
> any other Windows application. You don't even need to type a path -
> drag a folder onto the Command Prompt window and see what happens.
You mean that it opens the Command Prompt window and switches me to that
directory? Ho, hum.
Mike Rivers
> There is no lubricant, and nothing to cause stiction. But when
> turned off, the heads don't even sit on the disk. In Seagate drives, the
> heads sit on a stippled landing zone surface, textured by a laser. In
> Western Digital disks, the heads sit on low friction plastic loading ramp.
That's what I thought. It used to be that using a command to "park" the
heads
in a landing zone when turning off a computer to move it, but the ramp came
along pretty early. I remember that one of the drive manufacturers put some
code in the firmware that detected if the platter didn't start turning
when the
drive was powered up. It gave the head actuator motor a few high current
jolts to attempt to break it free.
About the only thing I can think of that could go wrong with a modern
disk drive
that's been in storage too long would be that lubricant on the head arm and
spindle bearings would dry out or become gummy.
Frank Stearns
>Frank Stearns wrote:
Right, but this is windows, where things aren't always logical! :) If you were in
explorer, do you want to print a folder list, or contents of a file? Explorer seems
to be clueless to the concept that you might ever want to print the list of files
you see in the right-hand pain (oops, sorry, I mean "pane").
Sometimes, a hardcopy file listing is a *very* useful thing to have; seems that
explorer ought to have a way to do this without fuss.
Also, it'd be great to at times simply copy files names (and other info) right out
of the right-hand pane and be able to paste them somewhere, say a text document....
Nope. Nothing happens at all, or all hell breaks loose if you try this -- windows
attempts to run the apps, or make them an embedded object of some kind in the target
app.
If you poke around the 'net, you'll find several third-party replacements for
explorer that do explorer's functions and many more. Some are even multiple-pane so
that you can see two different parts of your file system in the same app (I just
open as many explorers as I need). I seem to recall some that could even print a
file list.
Soundhaspriority
The lubricants they use are extremely sophisticated. The "oil" in a fluid
dynamic spindle bearing is intended to last 5 years at up to around 140F.
One drive engineer I've spoken to feels that the ultimate cause of demise is
particle buildup. Every time the head starts or stop, even with ramp
loading, some minute abrasion occurs. All this occurs on a scale that is so
minute, there's no intuition for it. But these days, the air cushion between
the head and the platter is only a few atoms!
Bob Morein
(310) 237-6511
Laurence Payne
wrote:
No, open the window THEN drop the folder in. In this case, it would
be a good idea to type "dir" first. Then all you have to do is type
in the switches.
Peter Larsen
> There are many flavors of backup software that will write
> compressed/proprietary save-set formats, but I learned years ago
> never to fully trust those.
Aamen!
> What I'd love is backup software that will simply write a directory
> structure onto the media of choice, each member file in its native
> format, with enough wits about it to prompt for the next disk as it
> goes, retaining the original directory structure as it goes.
Methinks the backup proggie that goes with the Maxtor OneTouch drives is
such a proggie,
I have a second hand Maxtor, perhaps it is still on it so that I can verify.
Xcopy properfly scripted is btw. also such a program. The differenc is that
it does not remove deleted files from the backup, that may be an advantage
or a disadvantage. On kessels.com you will find one of many
directory-syncronisation programs, you can also search microsoft.com for
synctoy.
> Frank
Kind regards
Peter Larsen
Richard Crowley
> That is correct. The fly height is so small, there is no room at all for a
> viscous layer. The surface is pure diamond.
That is all fine and well. But there are dozens of other things that can
(and DO) go wrong. Even hard drives that are powered up and
spinning fail. And those sitting inert on the shelf are even more prone
to failure.
Hard drives are NOT reliable enough for true "archive" application.
Virtually all of this planet's most valuable data is still archived on
digital mag tape.
Richard Crowley
> So you put it in a box. When Mackie came out with the HDR24/96, they had a
> package for the removable drive (a standard 3-1/2" hard drive in a mobile
> rack) that was an anti-static foam lined case the same size as
> a U-Matic videocassette. There was room for several sheets of paper, you
> could slip a cover sheet behind a protector on the outside of the case
> with the essential identification data on the front, back, and spine, and
> it would fit on a shelf like a book. And before you ask, I've never found
> those cases anywhere else but from Mackie. I tried using a
> U-Matic case, but it has bumps molded into the back to block the reels.
I use these...
http://www.wiebetech.com/products/cases.php
They are a bit fiddly and overpriced, but there's nothing else
currently available AFAIK. Seems like a great opportunity for
somebody to make "Tupperware-like" containers just the right
size for 3.5" (and 2") hard drives.
Soundhaspriority
"Richard Crowley" <rcro...@xp7rt.net> wrote in message
news:7nv67aF...@mid.individual.net...
> "Soundhaspriority" wrote...
>> That is correct. The fly height is so small, there is no room at all for
>> a viscous layer. The surface is pure diamond.
>
> That is all fine and well. But there are dozens of other things that can
> (and DO) go wrong. Even hard drives that are powered up and
> spinning fail. And those sitting inert on the shelf are even more prone
> to failure.
>
> Hard drives are NOT reliable enough for true "archive" application.
> Virtually all of this planet's most valuable data is still archived on
> digital mag tape.
>
You are correct that, individually, hard drives are not reliable enough. The
failure rate is around 3% per year. The solution I suggested to Frank is a
backup system with a human agent as an integral part. I performed a simple
multiplication to compute the probability of a joint failure. The number is
low, but it depends upon an efficient human agent to duplicate a drive as
soon as it fails.
Frank was after something involving DVDs. The whole idea behind DVDs is that
the human agent doesn't have to be nearly as vigilant. On the other hand, if
two DVDs come out of the same batch, and they have an unexpectedly high fade
rate, then when the human agent verifies them, there is a good chance both
have failed.
I think my suggestion of replacing DVDs with multiple hard drives is valid.
Tape is a different story. In the hands of professional archivists, tape is
considered the gold standard. But would it really work for an individual? I
have my own story about this. I recorded sound for a short film on about 12
DATs, using the HHB PortaDat, a 4 motor gold-standard field recorder. The
tension loop of this recorder was so well designed, the makers reported that
head replacements simply didn't occur.
The tapes were recorded in South Jersey in the humid summertime. Several
years later, I attempted to read them. Not a single tape was readable, on
the PortaDat, or an Aiwa. The tapes had shrunk in my dry basement. From my
point of view, I had lost them. The professional archivist, whose tapes are
stored in humidity controlled vaults, would not have had the problem. If he
had the problem,, he would have the resources to restore the tapes to
controlled humidity. Not me, and not Frank.
But there was a happy ending. I had all the wav files on DVDs, which were
still readable. Moral: different horses for different courses.
Bob Morein
(310) 237-6511
Mike Rivers
> I use these...
> http://www.wiebetech.com/products/cases.php
> They are a bit fiddly and overpriced, but there's nothing else
> currently available AFAIK. Seems like a great opportunity for
> somebody to make "Tupperware-like" containers just the right
> size for 3.5" (and 2") hard drives.
Those look more substantial and useful than the "pencil box" cases
that I've bought from Staples or Office Depot, which only serve to keep
your fingers off the drive and keep them from sliding off the shelf. They
used to sell for about a quarter around school opening time, but now
they seem to be around a buck and thinner than the ones I bought
8-10 years ago. But then most things are.
Richard Crowley
> "Richard Crowley"wrote ...
>> "Soundhaspriority" wrote...
>>> That is correct. The fly height is so small, there is no room at all for
>>> a viscous layer. The surface is pure diamond.
>>
>> That is all fine and well. But there are dozens of other things that can
>> (and DO) go wrong. Even hard drives that are powered up and
>> spinning fail. And those sitting inert on the shelf are even more prone
>> to failure.
>>
> I would like to see a link to the above assertion.
I shall post a photo of my stack of dead hard drives.
Or the barrels full of them back at the office.
>> Hard drives are NOT reliable enough for true "archive" application.
>> Virtually all of this planet's most valuable data is still archived on
>> digital mag tape.
>>
> You are correct that, individually, hard drives are not reliable enough.
> The failure rate is around 3% per year. The solution I suggested to Frank
> is a backup system with a human agent as an integral part. I performed a
> simple multiplication to compute the probability of a joint failure. The
> number is low, but it depends upon an efficient human agent to duplicate a
> drive as soon as it fails.
That makes the assumption that "fail" precludes the ability
to read the data. I'd like to see a link to THAT assertion.
By my definition "fail" means that you can't read data and
it is too late to duplicate at that point.
> Frank was after something involving DVDs. The whole idea behind DVDs is
> that the human agent doesn't have to be nearly as vigilant. On the other
> hand, if two DVDs come out of the same batch, and they have an
> unexpectedly high fade rate, then when the human agent verifies them,
> there is a good chance both have failed.
>
> I think my suggestion of replacing DVDs with multiple hard drives is
> valid. Tape is a different story. In the hands of professional archivists,
> tape is considered the gold standard. But would it really work for an
> individual? I have my own story about this. I recorded sound for a short
> film on about 12 DATs, using the HHB PortaDat, a 4 motor gold-standard
> field recorder. The tension loop of this recorder was so well designed,
> the makers reported that head replacements simply didn't occur.
Nothing is that "gold-standard". Particularly not any kind of helical-
scan small-scale format like DAT. Even a built-like-a-battleship Nagra
can go out of alignment.
Soundhaspriority
> "Soundhaspriority" wrote ...
>> "Richard Crowley"wrote ...
>>> "Soundhaspriority" wrote...
>>>> That is correct. The fly height is so small, there is no room at all
>>>> for a viscous layer. The surface is pure diamond.
>>>
>>> That is all fine and well. But there are dozens of other things that can
>>> (and DO) go wrong. Even hard drives that are powered up and
>>> spinning fail. And those sitting inert on the shelf are even more prone
>>> to failure.
>>>
>> I would like to see a link to the above assertion.
>
> I shall post a photo of my stack of dead hard drives.
> Or the barrels full of them back at the office.
>
to failure than drives in use.
>>> Hard drives are NOT reliable enough for true "archive" application.
>>> Virtually all of this planet's most valuable data is still archived on
>>> digital mag tape.
>>>
>> You are correct that, individually, hard drives are not reliable enough.
>> The failure rate is around 3% per year. The solution I suggested to
>> Frank is a backup system with a human agent as an integral part. I
>> performed a simple multiplication to compute the probability of a joint
>> failure. The number is low, but it depends upon an efficient human agent
>> to duplicate a drive as soon as it fails.
>
> That makes the assumption that "fail" precludes the ability
> to read the data. I'd like to see a link to THAT assertion.
> By my definition "fail" means that you can't read data and
> it is too late to duplicate at that point.
>
I agree. Please reread my post. Store on multiple drives; duplicate the
survivor in event of a failure.
[snip]
Bob Morein
(310) 237-6511
Richard Crowley
> The assertion I specifically question is that stored drives are more prone
> to failure than drives in use.
100% of my failed drives were discovered to be dead upon being
taken down from the shelf and re-connected for use. Virtually all
of them simply fail to show any signs of life at all. No spin, not even
a click.
Soundhaspriority
manufactured ~2001-2002 fail that way, but not since.
But we're just trading anecdotes. If this is an industry problem, there
should be substantial study of it, represented in the Google archives. I
can't make a decision based on your anecdotes, my anecdotes, or anybody
else's. Failure rates are usefully characterized by statistics.
Bob Morein
(310) 237-6511
Don Pearce
Soundhaspriority
"Don Pearce" <sp...@spam.com> wrote in message
news:4b20d8f6...@news.eternal-september.org...
I know that paper. It doesn't cover drives sitting on the shelf.
I'll tell you what has worked for me. I have a bunch of machines, and each
has several big hard disks. As the years go by, I add/swap in bigger ones. I
just replicate my data around these machines. I've lost hard disks, but I've
never lost archives, because they're stored in duplicate and triplicate. I
started doing this in 2003,when I saw that my piles of DVDs were going to
reach to the moon. And there was the problem, which Frank has mentioned, of
splitting across them, and organizing them. Everything is at my fingertips,
and replication is easy.
Bob Morein
(310) 237-6511
Mike Rivers
> 100% of my failed drives were discovered to be dead upon being
> taken down from the shelf and re-connected for use. Virtually all
> of them simply fail to show any signs of life at all. No spin, not even
> a click.
What "vintage" were these drives? Fifteen years ago (certainly a reasonable
archive lifetime) I wouldn't have advised using a "stored" hard drive
for backup.
Drives today are different. Given your "scare" I just took a couple of
drives
that haven't been spun up for five years and tried them. They all worked
fine.
You have more problems with things you discuss here than most people seem
to have. I won't attempt to explain that, just observe it.
Frank Stearns
>Richard Crowley wrote:
Been interesting to watch this thread.
For me the "background nag" in all this, especially as I trudge further along into
geezerhood, is just how quickly time goes by.
I could have sworn that project X was oh, say, just a few years ago, when it might
have been 10 years, 12, even more, in some cases. That certainly happens with the
more casual memories in our lives.
Older media, such as pre mid-70s analog tape, seems to survive nicely, if cared for,
and can (and has) been "forgotten" in the way just noted.
But will even a modern harddrive survive if not spun for, say, 30 years? And even if
the drive itself is sound (no pun), will there be some place to plug it into your
little optical computer the size of a sugar cube? What about "legacy" applications,
such as Protools 22? (Sure, start over with the DAW of the day -- or just let the AI
do it -- but it's good to look at some of those old details first, if you still
can.)
Will a company then akin to those today making a $49 turntable-to-USB make a cute
kit to talk to your "grandfather's computer drives"?
Seems to me that "passive" storage, such as CD or DVD, has a slight survivability
edge in this regard than does an "active" device, such as an HD, which requires a
lot of "external stuff" to bring it up make sense of the data.
I'm all for disbursement of data; I too have multiple generations of client
information moving from disk to disk (still haven't found ideal management software
for that, still looking). This is the best method of survival, including getting
pieces out to other locations, say DVDs or HDs, into the hands of clients.
OTOH, this is how culture filters junk. A stellar performance/recording is worth
active vigilance over the years. But another out-of-tune,
just-doesn't-quite-have-it/wanna-be star, perhaps would be best left on the shelf.
Oh, right, that's an awful lot of current pop music... Oops, sorry, showing my
cantankerous age. <w>
Soundhaspriority
it last longer. That doesn't answer the 30 year question. But unless DVDs
are very carefully stored, and even then, the dye instability might not let
them reach 30 years.
> Will a company then akin to those today making a $49 turntable-to-USB make
> a cute
> kit to talk to your "grandfather's computer drives"?
>
I wouldn't count on it.
> Seems to me that "passive" storage, such as CD or DVD, has a slight
> survivability
> edge in this regard than does an "active" device, such as an HD, which
> requires a
> lot of "external stuff" to bring it up make sense of the data.
>
Oh no. DVDs will not be in common use in 30 years. You'll be hunting "eBay"
for a working machine. If the dye in the disks hasn't decayed by then.
> I'm all for disbursement of data; I too have multiple generations of
> client
> information moving from disk to disk (still haven't found ideal management
> software
> for that, still looking). This is the best method of survival, including
> getting
> pieces out to other locations, say DVDs or HDs, into the hands of clients.
>
> OTOH, this is how culture filters junk. A stellar performance/recording is
> worth
> active vigilance over the years. But another out-of-tune,
> just-doesn't-quite-have-it/wanna-be star, perhaps would be best left on
> the shelf.
> Oh, right, that's an awful lot of current pop music... Oops, sorry,
> showing my
> cantankerous age. <w>
>
> Frank
> Mobile Audio
>
Yes, because any backup system intended for perpetuity must be tended by
archivists. There is an OT but analogous example: The Yucca Mountain nuclear
waste depository. How can people be prevented from digging into it 10,000
years or more from now. No sign would last, no language be readable. It is
actually part of the Yucca Mountain proposal to create a "priesthood", a
verbal perpetuation of a legend, to be the durable signage.
Bob Morein
(310) 237-6511
Mike Rivers
> For me the "background nag" in all this, especially as I trudge further along into
> geezerhood, is just how quickly time goes by.
>
> I could have sworn that project X was oh, say, just a few years ago, when it might
> have been 10 years, 12, even more, in some cases. That certainly happens with the
> more casual memories in our lives.
Getting old fast sure sucks, doesn't it? <g>
> Older media, such as pre mid-70s analog tape, seems to survive nicely, if cared for,
> and can (and has) been "forgotten" in the way just noted.
The risk with so many media types that have come and gone is not so much
the
degradation of the media, but the inability to play it back. Punch cards
that are kept dry
can last a long time, but do you know where your card reader is? How
about that Akai
A-DAM 12-track digital cassette? And do you have a computer with an
interface card
that will talk to an MDM or ST-506 hard drive, assuming you can get it
to spin?
There are, fortunately, guys like Adrian who can play cylinders with
only a little more
effort than playing CDs, but they're few and far between. Fortunately
analog tape still
has a cult following that's keeping at least some track formats alive.
How long will it
last?
> But will even a modern harddrive survive if not spun for, say, 30 years?
Nobody knows, because they haven't been around for 30 years yet, and
probably
won't ever be. Accelerated life tests on DAT suggested that they'd last
50 years yet
I know that there are a lot of folks who have DATs that won't play. That
doesn't
necessarily mean that they're unplayable, it could be that they have a
DAT deck
that needs maintenance that it's never had. But it means that they don't
have a quick
and ready archive.
> What about "legacy" applications, such as Protools 22?
I still have some Wordstar files around here. Fortunately there's enough
clear text
in them that I could re-created them (by hand) if I needed to. But back
then I kept
paper copies of everything I wrote on the computer, and I still have those.
> Will a company then akin to those today making a $49 turntable-to-USB make a cute
> kit to talk to your "grandfather's computer drives"?
I doubt it. Phonograph records are something that everybody has, or had,
and has
memories of. Hard drives are something that most people throw away with
the old
computer.
> Seems to me that "passive" storage, such as CD or DVD, has a slight survivability
> edge in this regard than does an "active" device, such as an HD, which requires a
> lot of "external stuff" to bring it up make sense of the data.
There's the mechanical considerations, and then there's the human ones,
too. CDs
have already survived 25 years, and while you can't buy a music CD
player any more
(they're all DVD players) you can at least buy a new box that can play
music CDs, and
you can still buy a computer drive that will read a data (or music) CD.
But SATA hard
drives have completely replaced parallel IDE drives and in a couple of
years you'll need to
adapt your old IDE drives to USB or something before you can read them.
At least that's
still an option. I'm glad I have a couple of plug-on IDE-USB adapters.
Hopefully USB2
backward compatibility will be with us for several more years. But I'll
bet there won't be an
IDE-USB3 adapter made.
> I'm all for disbursement of data; I too have multiple generations of client
> information moving from disk to disk (still haven't found ideal management software
> for that, still looking). This is the best method of survival, including getting
> pieces out to other locations, say DVDs or HDs, into the hands of clients.
>
> OTOH, this is how culture filters junk.
I've long ago stopped keeping source recordings that I've made. I'll
keep a few copies of
the final mix of a multitrack project, but after the client is satisfied
with the mix, it doesn't
hurt me to delete the files. If he wants a copy, I'll gladly make one
for him on whatever media
he wants, but I hope I don't have to see it again 25 years from now and
tell him "sorry, I don't
have any way to play that any more."
Les Cargill
So rotate three or four (I use two - I'm not that paranoid ) USB
drives at whatever frequency of backup you choose. CRC check them.
Run the CRC check whilst you're at work.
When one becomes untrustworthy, hie thee down to Best Buy and replace
it. The probability of two of them failing within the week
are astronomical, and you still have the live versions of what
you're backing up. At least I do.
These days, individual drives come in denominations of *terabytes*.
And they're about the cost of a nice meal.
You cannot trust them, but you can create a scenario where you
mostly *can*.
--
Les Cargill
PStamler
> I've long ago stopped keeping source recordings that I've made. I'll
> keep a few copies of
> the final mix of a multitrack project, but after the client is satisfied
> with the mix, it doesn't
> hurt me to delete the files. If he wants a copy, I'll gladly make one
> for him on whatever media
> he wants, but I hope I don't have to see it again 25 years from now and
> tell him "sorry, I don't
> have any way to play that any more."
But I bet Steve Puntolillo will.
Peace,
Paul
Richard Crowley
> You have more problems with things you discuss here than most people seem
> to have. I won't attempt to explain that, just observe it.
I didn't say that 100% of my drives on the shelf failed.
I said that of the drives that failed, 100% of them failed
with the "DOA" symptom. I haven't calculated the % of
failures, but it is high enough (maybe 5%) to make me
abandon any notion of hard drives being "archival".
Richard Crowley
> What "vintage" were these drives? Fifteen years ago (certainly a
> reasonable
> archive lifetime) I wouldn't have advised using a "stored" hard drive for
> backup.
> Drives today are different. Given your "scare" I just took a couple of
> drives
> that haven't been spun up for five years and tried them. They all worked
> fine.
It seems true that the ones that fail are the very old ones.
(5-10 years old). But of course we have no experimental
data on how new drives will fare when then are "old". :-)
Scott Dorsey
If the interface can't be interrogated by the computer, try swapping boards
with another disk drive of the same model.
There are a lot of failures of this sort of thing that are the result of
the move to lead-free solder.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."
Jason Warren
says...
>
> There are a lot of failures of this sort of thing that are the result of
> the move to lead-free solder.
> --scott
Amen! Tin whiskers are the bane of my existence. Just when I
thought I had SM soldering figured out I find that a solder
joint that's good on Monday may magically fail a month
later. The main computer onboard the ISS failed for the same
reason. The Russians who built it ignored the fact that the
spec stipulated NOT to adhere to RoHS.
Scott Dorsey
>In article <hfmqu6$fg9$1...@panix2.panix.com>, klu...@panix.com
>says...
>
>> There are a lot of failures of this sort of thing that are the result of
>> the move to lead-free solder.
>
>thought I had SM soldering figured out I find that a solder
>joint that's good on Monday may magically fail a month
>later. The main computer onboard the ISS failed for the same
>reason. The Russians who built it ignored the fact that the
>spec stipulated NOT to adhere to RoHS.
That's not a tin whisker issue so much as a cold joint issue, but both
are bigtime problems with the new lead-free solders.
This means that consumer electronics now have an even shorter lifespan,
causing _more_ electronic junk in landfills rather than less. Which defeats
the whole purpose of the thing.