View Issue Details

IDProjectCategoryView StatusLast Update
0005152CentOS-6kernelpublic2011-09-28 11:01
Reporterjegunn 
PriorityurgentSeveritycrashReproducibilityrandom
Status newResolutionopen 
PlatformASUS M4A88-TDV, AMD 1055TOSCentOSOS Version6.0
Product Version6.0 
Target VersionFixed in Version 
Summary0005152: Data loss on Corsair F120 SSD after crash
DescriptionSystem crashes randomly. After crash, all data from last ~20 hours on SSD, which
contains /boot/, /, swap, /home, and a data partition, is gone. Nothing in any
log. have noatime,discard,defaults as options in fstab for all partitions on SSD. The disk has been in use with no incident under F13 for the past two years.

My general impression (4 machines, 2 years, ZERO groubis that F13 is a much more solid distro than Centos6 (see previous bug for vmware).
Steps To ReproduceThe crashes happen randomly, seemingly always at times when the machine is idle.
(it has only happened twice, but data loss is not something I want to have to
get used to.)
Additional InformationI can find nothing about this trouble in your material or in RHEL6 material,
but it is devastating. It is NOT the disk, because it is well used and proven, but everything else is new. I gather that the trim capability is natively in the 2.6.32 kernel, but I ran with 2.6.32 under 64-bit F13 for a long time
before they went to 2.6.34 and had no trouble. I have now turned off discard
and will see what happens.
TagsNo tags attached.

Activities

athmane

athmane

2011-09-28 11:01

developer   ~0013414

Can you attach /var/log/messages , /var/log/dmesg (or dmesg output), fsck output and any oops/panic you see on the console.

Issue History

Date Modified Username Field Change
2011-09-28 00:56 jegunn New Issue
2011-09-28 11:01 athmane Note Added: 0013414