Gluster users - Nov 2017 - file shred

Hi,

if we were to use shred to delete a file on a gluster volume, will the
correct blocks be overwritten on the bricks?

(still using Gluster 3.6.3 as have been too cautious to upgrade a
mission critical live system).

Cheers,
Kingsley.

On 11/08/2017 11:36 PM, Kingsley Tart wrote:
> Hi,
>
> if we were to use shred to delete a file on a gluster volume, will the
> correct blocks be overwritten on the bricks?
>
> (still using Gluster 3.6.3 as have been too cautious to upgrade a
> mission critical live system).
When I strace `shred filename`, it just seems to write + fsync random 
values into the file based on some seed from /dev/urandom. Since these 
are normal syscalls, and since gluster just sends these FOPS to the 
bricks, it should work fine.
All caveats listed in `man shred` for the on-disk file system (XFS) 
still apply.
Regards,
Ravi
> Cheers,
> Kingsley.
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://lists.gluster.org/mailman/listinfo/gluster-users

> When I strace `shred filename`, it just seems to write + fsync random 
> values into the file based on some seed from /dev/urandom. Since these 
> are normal syscalls, and since gluster just sends these FOPS to the 
> bricks, it should work fine.
> All caveats listed in `man shred` for the on-disk file system (XFS) 
> still apply.
Thanks Ravi.

-- 
Cheers,
Kingsley.