![](/i/fill.gif) |
![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Invisible wrote:
> If stuff costs money then fine. But don't claim something is free when
> it isn't. That's deception.
That's "Search Engine Optimization". :-)
--
Darren New / San Diego, CA, USA (PST)
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
>>> Yeah, I can see how that might corrupt the filesystem. But badly
>>> enough to render it unreadable? You would think you'd just get a few
>>> filesystem inconsistencies. The error message sounds more like the
>>> superblock is unreadable or something...
>>>
>>
>> Depends on what the computer is doing when the stick is pulled out.
>> Think what could happen, if the computer were writing file allocation
>> tables...
>
> Or, say, the superblock. ;-)
Yeah, but AFAIK you only write to the superblock when you initially
format the volume?
I can understand a file or folder being broken, but the entire FS?
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
>> If stuff costs money then fine. But don't claim something is free when
>> it isn't. That's deception.
>
> That's "Search Engine Optimization". :-)
That's what I said. Deception. :-P
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Invisible wrote:
> Yeah, but AFAIK you only write to the superblock when you initially
> format the volume?
Hmmm... On FAT? I'm not sure. There might be a flag there that marks
whether the drive was dismounted cleanly or not. Or if you screw up the
root directory, I can see that mucking things up too. If there were just
a couple of directories, and the first sector of the root directory got
scrambled because she pulled it out while it was writing, I can see that
messing things up.
> I can understand a file or folder being broken, but the entire FS?
FAT usually has two FATs, too, for just such a reason. It's possible she
got a warning that the first was broken and just ignored it until the
second got messed up too.
I'm not real sure of the electrical characteristics of a USB stick such
that it gets the whole chip messed up by being pulled out while writing,
but I have to believe avoiding that was one of the design goals.
--
Darren New / San Diego, CA, USA (PST)
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Invisible wrote:
>>> If stuff costs money then fine. But don't claim something is free
>>> when it isn't. That's deception.
>>
>> That's "Search Engine Optimization". :-)
>
> That's what I said. Deception. :-P
I'm not disagreeing. :-)
I was told by someone at google that the two highest cost ads are for
"data recovery" and for "mesothelioma" (the cancer you get from asbestos).
--
Darren New / San Diego, CA, USA (PST)
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
On Wed, 27 Aug 2008 08:57:11 +0100, Invisible wrote:
> Jim Henderson wrote:
>
>> I had asked questions about using a journaling filesystem and it was
>> told. Yes, I could've read the specs but I didn't. So shoot me.
>
> Wear-leveling, anyone?
Probably contributes to the increased life of the devices these days.
Jim
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Darren New wrote:
> Invisible wrote:
>> Yeah, but AFAIK you only write to the superblock when you initially
>> format the volume?
>
> Hmmm... On FAT? I'm not sure. There might be a flag there that marks
> whether the drive was dismounted cleanly or not. Or if you screw up the
> root directory, I can see that mucking things up too. If there were just
> a couple of directories, and the first sector of the root directory got
> scrambled because she pulled it out while it was writing, I can see that
> messing things up.
Yeah. Well it's not complaining the filesystem is broken, it claims that
there *is* not filesystem at all. That's what I find so puzzling...
>> I can understand a file or folder being broken, but the entire FS?
>
> FAT usually has two FATs, too, for just such a reason. It's possible she
> got a warning that the first was broken and just ignored it until the
> second got messed up too.
>
> I'm not real sure of the electrical characteristics of a USB stick such
> that it gets the whole chip messed up by being pulled out while writing,
> but I have to believe avoiding that was one of the design goals.
As I understand it, to write to flash RAM, you erase a small block and
then completely re-flash it. If that got interrupted, your whole block
is gone.
No idea how large a flash block is - I'll bet it isn't the same size as
a filesystem block! ;-)
You'd think a tiny little capacitor would keep the electronics running
for enough split seconds to shutdown vaguely cleanly... *shrugs*
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Orchid XP v8 wrote:
> Yeah. Well it's not complaining the filesystem is broken, it claims that
> there *is* not filesystem at all. That's what I find so puzzling...
Right. The superblock (i.e., the boot block) holds a table that says
where the file system is, how many blocks are in it, which block the
root directory starts on, and so on. If that got wiped, then no, you
don't have any file system. :)
> No idea how large a flash block is - I'll bet it isn't the same size as
> a filesystem block! ;-)
I imagine it depends.
> You'd think a tiny little capacitor would keep the electronics running
> for enough split seconds to shutdown vaguely cleanly... *shrugs*
Not when you're writing flash. That tends to take a lot of current.
--
Darren New / San Diego, CA, USA (PST)
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
>> Yeah. Well it's not complaining the filesystem is broken, it claims
>> that there *is* not filesystem at all. That's what I find so puzzling...
>
> Right. The superblock (i.e., the boot block) holds a table that says
> where the file system is, how many blocks are in it, which block the
> root directory starts on, and so on. If that got wiped, then no, you
> don't have any file system. :)
I can see how that would be. What I don't see is why the hell you would
ever be writing data to anywhere near this location...
That again, I have no idea how FAT works.
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
Orchid XP v8 wrote:
> I can see how that would be. What I don't see is why the hell you would
> ever be writing data to anywhere near this location...
Flash isn't like a disk drive. It's electronic. If you zap the
electronics by (for example) pulling it out while it's writing, I don't
know what it would mess up.
And, as I said, there *is* a flag somewhere, I think, that says whether
the disk needs to be checked when remounted. I *think* that's in the
root block.
--
Darren New / San Diego, CA, USA (PST)
Post a reply to this message
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |
| ![](/i/fill.gif) |
|
![](/i/fill.gif) |
|
![](/i/fill.gif) |
| ![](/i/fill.gif) |