You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Best to leave LV creation out of scope (VG validation, size specification, availability checking, filesystem selection and create-time filesystem options, etc.)
LV is mounted temporarily if required (normal mount is expected to correctly detect the filesystem type)
Root filesystem cache is copied there instead of the subdir
The text was updated successfully, but these errors were encountered:
You most likely want to handle brtfs volumes as well for btrfs aficionados :)
If you are a zfs fan, this is mostly something to handle as well and should be similarly done.
Last but not least, one thing that current lxc-create does not handle is lvm thin volume or whatever is the feature again to create a minimal image to bootstrap your new volume with copy-on-write internally which should speed things up a fair bit.
I can see autodetection of installed filesystems with /proc/filesystems ...
I worry that some of them and advanced users will require create-time options that will cause the naive approach of mkfs.${FILESYSTEM} /dev/target to fail
Thinking of creating mkfs.${FILESYSTEM} as a default command line, prompting the user to confirm (ie. they can add custom options, have a script handle free space calculations, quotas, advanced creation logic, etc.)
Any recommended defaults that differ from mkfs.${FILESYSTEM} could be rolled in to lxc-gentoo for filesystems on which that is always a bad idea
Snapshot and thin provisioning support segmented to issue #42.
Rough thinking at present is:
mount
is expected to correctly detect the filesystem type)The text was updated successfully, but these errors were encountered: