really strange ext4 behavior
sandeen at redhat.com
Mon Feb 14 23:21:15 UTC 2011
On 2/14/11 2:14 PM, Michał Piotrowski wrote:
> W dniu 14 lutego 2011 20:47 użytkownik Eric Sandeen
> <sandeen at redhat.com> napisał:
>> On 2/13/11 12:29 PM, Dennis Jacobfeuerborn wrote:
>>> On 02/12/2011 11:52 PM, Ric Wheeler wrote:
>>>> On 02/12/2011 05:31 PM, Michał Piotrowski wrote:
>>> If ext3 was running fine without barriers for all these years why is this
>>> such a problem with ext4? Does ext4 do something differently that barriers
>>> are now required?
>> barriers are always required for integrity if you have a volatile write cache;
>> this is true for ext3 as well. You may not see problems on every power loss,
>> but eventually you will.
>> The problems are often found after the fact, with a subsequent runtime or fsck
>> error, so the culprit may not be immediately obvious.
> What are the recommended "best practices" for mounting ext3/4 file system?
> For performance - noatime, for those who care about data integrity -
> "barrier=1,data=journal" or just "barrier=1" if we also care about
> performance. Am I missing something?
There is no real best-practice tuning without workload details;
without that, "defaults" is best practice. :)
(except for ext3, where, for data integrity with a volatile writeback
cache, defaults + barriers=1, since that safe default was never accepted upstream)
More information about the devel