[SGVLUG] SUSE abandoning ReiserFS

[Mike Fedyk]

Dustin Laurence wrote:
> On Mon, Oct 02, 2006 at 05:38:02PM -0700, Emerson, Tom wrote:
>   
>> The "one thing" I really liked was resizing-on-the-fly -- does ext(n) or
>> XFS/JFS support this as well?  [or is this actually an artifact of using
>> volume groups and LVM?]
>>     
>
> $ man resize2fs
>
> [[snip]]
>
>        If the filesystem is mounted, it can be used to expand the size of  the
>        mounted filesystem, assuming the kernel supports on-line resizing.  (As
>        of this writing, the Linux  2.6  kernel  supports  on-line resize  for
>        filesystems mounted using ext3 only.).
>
> So that means it mostly can now do what reiser can do, I think--grow but
> not shrink online.

Don't forget that if you expect to grow an old (created with tools that 
didn't support online filesystem size growth) ext3 filesystem beyond 
certain thresholds it will need to be prepared offiline.

Does reiser3 shrink online?  I know it can grow online.

And as for the other filesystems, jfs can grow on/offiline and shrink 
offline, and xfs can only grow on/offline but not shrink which is a 
weird restriction and seems to only be so because a tool to do so hasn't 
been written yet, but I could be wrong of course.

And as for growth and shrinking being a property of LVM, it isn't.  LVM 
just creates an environment where that capability is useful more often. 

Also another surprising thing is that managing growing/shrinking LVM 
logical volumes and their contained filesystem is a manual process.  I 
typically resize the filesystem a bit smaller than I expect my new 
volume size to be (when shrinking) and then run the filesystem resize 
utility without a size specification to use the remaining portion of the 
newly sized logical volume.  IMO this should be automated and 
standardized by the LVM scripts, but isn't.

Has anyone looked into this?