[ale] Centos

Michael Trausch mike at trausch.us
Mon Jul 28 14:48:49 EDT 2014


Don't know that it works on ZFS either. It makes little sense with snapshot volumes in LVM/LVM2 as well. For LVM you want to use lvdisplay and for btrfs you want "btrfs fi du".

Sent from my iPhone

> On Jul 28, 2014, at 10:45 AM, Preston <preston.lists at gmail.com> wrote:
> 
>> On 7/25/2014 7:58 PM, Michael Trausch wrote:
>> I will note that btrfs is no longer "beta" and is now used in production environments far larger and with far more robust requirements than anything I have had to build for a client, for sure.
>> 
>> The data structures are sound and it scales much better than ext* for volumes > 2 TB, when you need more than about 20,000 entries in a directory, or when you need to not mess with complex middlewares between FS and block device for any one of a number of reasons. It provides better robustness than LVM or mdraid in many situations too, because all redundancy assurances are per object, not per block device.
>> 
>> Doesn't reduce the need for a quality backup regimen, though. Nothing ever will.
> 
> just a quick note, I didn't realize that my old "df -h" command wouldn't
> list the correct drive usage while running a btrfs filesystem. glad it
> was pointed out to me before I really started throwing stuff at it.
> 
> Preston
> 
> -- 
> Be who you are and say what you feel, because those who mind don’t matter and those who matter don’t mind.
> -Dr. Seuss
> 
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://mail.ale.org/mailman/listinfo/ale
> See JOBS, ANNOUNCE and SCHOOLS lists at
> http://mail.ale.org/mailman/listinfo



More information about the Ale mailing list