lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Message-ID: <7642a14c-67bf-4d25-b54e-419c7a650330@hogyros.de>
Date: Wed, 21 Aug 2024 18:38:52 +0900
From: Simon Richter <Simon.Richter@...yros.de>
To: linux-ext4@...r.kernel.org
Subject: Overwriteable media errors and data recovery

Hi,

my underlying block device lost a few blocks due to media errors, but 
the blocks involved are not permanently bad, so the badblocks mechanism 
isn't entirely the correct tool here.

Is there a good flow to recover from *past* media errors and get a list 
of affected files?

Use cases:

  - Harddisk reports read errors, but remaps blocks on write
  - Volume manager reports read errors because all mirrors were lost, 
but spares exist
  - ddrescue copied media, with a list of blocks that could not be copied

Basically, what I'd like to do is go from a damaged file system and a 
list of damaged blocks to a good file system and a list of damaged files 
(which allows me to check if I can restore them from backup, or, if they 
came from a package, simply reinstall them).

Could there be something like a file attribute "damaged" that is set by 
fsck, and that I can scan for inside a mounted file system?

    Simon

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ