On 9/7/23 21:55, David Sterba wrote:
On Wed, Sep 06, 2023 at 05:49:05PM +0800, Anand Jain wrote:
On 9/6/23 04:23, Guilherme G. Piccoli wrote:
On 05/09/2023 13:50, David Sterba wrote:
[...]
I'd like to pick this as a feature for 6.7, it's extending code we
already have for metadata_uuid so this is a low risk feature. The only
problem I see for now is the name, using the word 'single'.
We have single as a block group profile name and a filesystem can exist
on a single device too, this is would be confusing when referring to it.
Single-dev can be a working name but for a final release we should
really try to pick something more unique. I don't have a suggestion for
now.
The plan for now is that I'll add the patch to a topic branch and add it
to for-next so it could be tested but there might be some updates still
needed. Either as changes to this patch or as separate patches, that
depends.
Hi David, thanks for your feedback! I agree with you that this name is a
bit confusing, we can easily change that! How about virtual-fsid?
I confess I'm not the best (by far!) to name stuff, so I'll be glad to
follow a suggestion from anyone here heheh
This feature might also be expanded to support multiple devices, so
removing 'single' makes sense.
I'm not sure how this would work. In case of the single device we can be
sure which device belongs to the filesystem, just need the incompat bit
and internal uuid to distinguish it from the others.
With multiple devices how could we track which belong to the same
filesystem? This is the same problem we already have with scanning and
duplicating block devices.
The only thing I see is to specify the devices as mount options,
possibly with the bit marking the devices as seen but not
scanned/registered and never considered for the automatic mount.
Indeed, users have no means to accurately identify and group the devices
unless this information is maintained in a configuration file (similar
to md? I think). What I mean to convey is that it's possible through
alternative methods with certain trade-offs. Personally, I don't prefer
the configuration file method, but not all use cases share the same
preference.
virtual-fsid is good.
or
random-fsid
I'm thinking about something that would be closer to how the devices'
uuids can be duplicated, so cloned_fsid or duplicate_fsid/dup_fsid.
Virtual can be anything, random sounds too random.
At each mount, the fsid will be different and temporary; it may or may
not be a duplicate device. So, I believe temp_fsid, proxy_fsid, or
virtual_fsid could represent these configurations.
Thanks, Anand
