← Back to team overview

kernel-packages team mailing list archive

[Bug 1507618] Re: Potential raid0 data corruption when using trim

 

** Package changed: ubuntu => linux-meta (Ubuntu)

-- 
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux-meta in Ubuntu.
https://bugs.launchpad.net/bugs/1507618

Title:
  Potential raid0 data corruption when using trim

Status in linux-meta package in Ubuntu:
  New

Bug description:
  There's a scary issue with fstrim on raid0 and raid10 mdadm devices.
  It is fixed upstream in linux/4.1.5-1, linux/3.16.7-ckt17-1, but I
  can't find anything in the changelogs for ubuntu.

  Here's the links:

  https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=793326
  http://www.spinics.net/lists/raid/msg49440.html

  I did my amateur analysis by extracting sources for current trusty
  kernel 3.13.0-65-generic and current trusty-lts-utopic
  3.16.0-50-generic and looking at files referenced in patches from the
  Martin K. Petersen's email, and it appears that 3.16 already has this
  fix included, and 3.13 is a completely different code base, so it
  isn't clear whether it is affected or not.

  I've disabled fstrim in cron until I understand the situation better.

  Could someone knowledgeable chime in on the status of this on Ubuntu,
  specifically Trusty and after?

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-meta/+bug/1507618/+subscriptions