Any one have a recommendation for backing up three VPS and a 2tb nas?
Glacier is too expensive for retrieval. Back blaze good?
On 2016-10-24 08:58, Carruth, Rusty wrote:Yes, mount goes through all the block devices probably starting with the first SCSI disk. If it's looking for a UUID and finds it on /dev/sda3, that's the one it'll use, even if the same UUID is on /dev/sdb1 . I think. IIRC, the label detection code in mount did that the last time I looked at it.
NOTE! WARNING! BEWARE!!! DD will almost certainly copy the UUID from
the source partition to the destination partition! I do NOT know what
havoc will result when linux looks for that UUID and finds 2…. (I’d
guess it takes the first one it finds
If you know you want to have the same UUID on 2 filesystems, you can use "dumpe2fs -h /dev/sdNN | grep UUID" , then pass the big hex string to the -U option of mkfs when you're making the new filesystem. Or the -U option of tune2fs if you've already done mkfs and copied stuff.
(I know about the UUID copy because I do that here at work all the
time. In my case, it’s a feature. In your case, it’s a bug)
IMHO, using filesystem labels is preferable to using UUIDs in /etc/fstab . Labels can be made short and meaningful to humans, while UUIDs really can't. (OK, -U feedface-dead-beef-0000-123456789abc works, but is silly.) Distros probably go the UUID route because it's generally easy to assume that UUIDs are unique, while filesystem labels may not be.
--
Crow202 Blog: http://crow202.org/wordpress
There is no Darkness in Eternity
But only Light too dim for us to see.
---------------------------------------------------
PLUG-discuss mailing list - PLUG-discuss@lists.phxlinux.org
To subscribe, unsubscribe, or to change your mail settings:
http://lists.phxlinux.org/mailman/listinfo/plug-discuss