On 03/19/2015 07:15 PM, Chris Murphy wrote:
On Thu, Mar 19, 2015 at 7:49 PM, don fisher <hdf3@xxxxxxxxxxx> wrote:
My rot
is mounted at /usr11. I tried:
UUID=26d63d99-92f4-4a49-878b-236f5e88af69 /usr11/srv btrfs
subvol=usr11/srv 0 0
but /usr11/srv did not mount. I replaced the number and it did mount. Would
like to use string like you suggested since they are more understandable. Do
you see what I did wrong?
Use 'btrfs subvolume list -a' like I suggested and use that path minus
any leading /. Chances are the path to the subvolume srv is just srv,
in which case it should be subvol=srv
Any thoughts on stability of the IDs across boots?
They're completely reliable between reboots and renames. They're
assigned at creation, and either btrfs sub create or btfs sub snap
creates subvolumes.
Thanks. I will try that. How does the mount distinguish between /srv and
/usr11/srv if the subvol=srv is the same for both? It appears to work,
for when I do a df the sizes are different. I guess it can tell from the
UUID of / as opposed to that of /usr11. I missed that one completely.
/dev/sdc2 20972544 6117628 14641012 30% /srv
/dev/sdb2 20972544 9492892 11112580 47% /usr11/srv
I would still like to know why we need to mount the subvolumes at all. I
tried to dismount one of the subvolumes mounted on / and got a:
umount: /var/log: target is busy
message, so could not test to /var/log access without a mount. It is
easy to access /usr11/var/log without it being mounted. Again, why mount
it? I understood mounting partitions, because that was the only access.
But the appears to be two access channels here.
I made another openSuse system on a 2.5" USB drive. I will use it for
the test tomorrow when more of my neurons are firing.
Thanks
Don
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
