0

Please see What happens in a raid (5) when one drive or more gets a bad sector?. I have a few .jpg files (and much more serious data). Some of the .jpg files become corrupted. I can tell because they are unviewable with viewer "geeqie". I am currently using btrfs on a crucial CT512MX1 ssd bought JAN 2015.

  1. (question 1) What maintenance you can do on btrfs instead of "fsck"?
  2. (question 2) What are the software/hardware requirements for solid state disks in raid 1 to accomplish this?" e.g. must it be able to read SMART data for ssd?

Here is a discussion which is too complicated: https://linux-btrfs.vger.kernel.narkive.com/BQK0Pz9w/how-does-btrfs-handle-bad-blocks-in-raid1

2
  • 1
    First, keep backups!, second use data scrubbing. (Reading all data of the drive periodically)
    – NiKiZe
    Nov 8, 2021 at 3:34
  • 1
    Please limit yourself to a single, specific question that can be answered. Right now there's a lot of rambling, possibly several questions, and nothing very clear. Nov 8, 2021 at 3:40

2 Answers 2

0

Yes, the answer to Question 1 is: use btrfs scrub to detect and fix damaged files: see

https://wiki.archlinux.org/title/Identify_damaged_files

specifically

"Unlike other filesystem types, btrfs has native support for reporting on damaged files. When scrubbing a partition, btrfs reads all data and metadata blocks and verifies checksums. It automatically repairs corrupted blocks if there is a correct copy available in a RAID configuration. btrfs also reports any unreadable sector along with the affected file via system log. "

1
  • 1
    As it’s currently written, your answer is unclear. Please edit to add additional details that will help others understand how this addresses the question asked. You can find more information on how to write good answers in the help center.
    – Community Bot
    Nov 8, 2021 at 4:29
0

You said « I am currently using btrfs on a crucial CT512MX1 ssd ». A single drive can’t be considered RAID1. Then btrfs will get you many advantages, but none about redundancy. If a sector can’t be read, the file is lost (more accurately can’t be read completely). The only advantage of btrfs here is that it stores a checksum of each block and will prefer to raise an I/O error instead of sending a incorrect value for some sectors : you can’t read the file instead of your application breaks because of a corrupted file.

2
  • yes I know it is not raid1 now; I want to know if any specific hardware is required (to purchase in future) if I want to implement the raid 1 scheme as in the answer below. or just it only necessary to get a pair of ssds for which smartd works? For example do you need disks designed for NAS or raid or enterprise ssd instead of any desktop ssd? Nov 8, 2021 at 21:24
  • Any pair of SSD can constitute a RAID 1 partition or redondant btrfs partition. NAS or enterprise drives are equal about it, but have generally a lower bit error rate (probability of error while reading). I don’t really know about SSD, but they are not all equal about the number of rewrites they support. Nov 8, 2021 at 21:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.