[Gluster-users] After gluster clean up sub directories becomes invisible

[Vijay Bellur]

On 06/12/2017 12:54 PM, Sangeeta Ramapure wrote:
> Steps to do gluster clean up
>
>
>
> 1.       umount -f /export/home/ecmsftp
>
> 2.       Remove the /export/home/ecmsftp mount point line from
> /etc/fstab file if it exists.
>
> 3.       Delete gluster file system volume if it exists; ignore if it
> does not exist.
>
> # gluster
>
> gluster>volume list
>
> eftpVol
>
> gluster> volume stop eftpVol
>
> Stopping volume will make its data inaccessible. Do you want to
> continue? (y/n) y
>
> volume stop: eftpVol: success
>
> gluster> volume delete eftpVol
>
> 4.       Deleting volume will erase all information about the volume. Do
> you want to continue? (y/n) y
>
> volume delete: eftpVol: success
>
> gluster> peer status
>
> Number of peers: 0
>
> 5.       Remove gluster-related configurations.
>
> brick_path=/gluster/eftpbrick
>
> [ -d $brick_path ] && setfattr -x trusted.glusterfs.volume-id $brick_path
>
> [ -d $brick_path ] && setfattr -x trusted.gfid $brick_path
>
> [ -d $brick_path/.glusterfs ] && rm -rf $brick_path/.glusterfs


Can you avoid removing $brick_path/.glusterfs? That should help in 
preserving essential metadata needed for gluster and allow you to access 
data from the mount point after volume is re-configured.

Alternately if you remove .glusterfs, you would need to run "find 
/mnt/point | xargs stat" or something similar to trigger lookups on all 
directories and files to recreate the metadata in .glusterfs.

Regards,
Vijay