Computing desk
< July 14	<< Jun \| July \| Aug >>	July 16 >

Welcome to the Wikipedia Computing Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

July 15

File-system, external storage, unsafe removal

Do different file-systems behave differently, when an external storage medium is removed unsafely? That is, it's connected through USB and I just pull it. Would something like ext3 be more reliable than NTFS? Hofhof (talk) 00:43, 15 July 2016 (UTC)[reply]

It depends on what features the filesystem has. ext3 and NTFS have rough feature parity—both are journaling file systems—so between those two there's unlikely to be any significant difference. Other features like copy on write (not present in those two file systems) can also prevent filesystem corruption in some situations. It should go without saying that you should not make a habit of unsafely removing storage devices, and no file system features are a substitute for backups. --71.110.8.102 (talk) 02:32, 15 July 2016 (UTC)[reply]

Aging of flash memory is caused by writing. So flash memory devices are written as less as possible. Removing the unsafe way may cause a lost of recent updates of the journal. Depending on the file system, when the journal is redundant, it is more robust due copies can be compared to the file it points and the checksum. --Hans Haase (有问题吗) 06:11, 15 July 2016 (UTC)[reply]

Here is an interesting (if somewhat old) overview of the journaling behavior of various file systems. ext3 has three journaling modes, and I gather that NTFS's journaling is more or less the same as ext3's middle level ("ordered"). What this means is that you won't end up with a broken filesystem that needs fsck/chkdsk, and you won't end up with random crud (or zeroes) in a file that was never written to it, but you may end up with out-of-order overwrites. That is, if you overwrote data chunks A, B, and C in that order, then yanked the drive too early, you might find that A and C (or even parts of them) made it to the disk and B didn't.

Note that these guarantees apply to disks that support some measure of data integrity in the first place. I've heard horror stories of SSDs being irreversibly damaged by sudden power loss (see [1]). That's reason enough to stop the drive first, regardless of the file system. I think HDDs are much better in this regard, even the cheaper consumer ones, but repeated sudden power loss may cause long-term damage to them too. -- BenRG (talk) 06:37, 15 July 2016 (UTC)[reply]

See SandForce and Write amplification to get an idea how this is working. A typical horror scenarior might be caused when the block is being relocated due its number of rewrites exceeded the maximum specified. If this relocating is a map as well, the question must be asked, where is its repair procedure tool? --Hans Haase (有问题吗) 06:54, 15 July 2016 (UTC)[reply]

Keybboard

My keyboard at some point completely ceased to function, and I am now using the super-tedious "on-screen keyboard". I likely bumped or hit something that caused this. My sleep-deprived brain is not working well, but I'm assuming this is a fairly simple issue. Any suggestions? Joefromrandb (talk) 02:04, 15 July 2016 (UTC)[reply]

Is that a laptop or desktop? Hofhof (talk) 02:20, 15 July 2016 (UTC)[reply]

What operating system are you using? Is this a wireless or wired keyboard? Have you checked that it's plugged in? If it's wireless, does it have working batteries? Have you tried restarting the system? --71.110.8.102 (talk) 02:34, 15 July 2016 (UTC)[reply]

It's a laptop. Yes, it's plugged in, and yes I've restarted it. I'm using Windows 10, if that's what you mean. Joefromrandb (talk) 02:45, 15 July 2016 (UTC)[reply]

Try to reinstall it using the add hardware wizard to install the driver package. Otherwise, you might need to replace the keyboards. In some laptops is rather easy to remove it. You could also try to remove it, reboot, and to connect it again.Hofhof (talk) 05:28, 15 July 2016 (UTC)[reply]

The most common cause of this I've seen is a stuck key/dirt in the keys. It fails a self test at startup and then it's disabled (using Windows, anyway). Just shut off the computer, run your hand back and forth over it a few times so all the keys clatter, and see if you have a keyboard. Wnt (talk) 10:17, 15 July 2016 (UTC)[reply]

If your laptop has a removable battery try this: Shut down the laptop, remove the power supply, remove the battery. Wait 1 minute, reinsert the battery, reinsert the power, and start the laptop again. Even when shutdown, laptops often keep the glidepad and keyboard powered up. Doing the above procedure will reset their microcontrollers. Otherwise, you can use an external USB keyboard for the laptop. LongHairedFop (talk) 11:36, 15 July 2016 (UTC)[reply]

Thank you both!! Wnt's suggestion worked like a charm. Joefromrandb (talk) 03:39, 17 July 2016 (UTC)[reply]

Joefromrandb, see Wikipedia:Reference_desk/Archives/Computing/2016_April_22#Throwing_orange_juice_on_a_laptop. I guess one of the keys is cut short, permanently pressed or two lines of the matrx keyboard, each computer keyboard today is, are cut short. --Hans Haase (有问题吗) 08:30, 18 July 2016 (UTC)[reply]