Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 24 Apr 2021 05:47:45 +0200
From:      Peter Libassi <peter@libassi.se>
To:        d@delphij.net
Cc:        mike tancsa <mike@sentex.net>, FreeBSD-STABLE Mailing List <freebsd-stable@freebsd.org>, Xin Li <delphij@delphij.net>
Subject:   Re: zfs native encryption best practices on RELENG13
Message-ID:  <4CFAA2E3-F8B0-41F3-BA2D-4802FC138E8C@libassi.se>
In-Reply-To: <56a4a35f-b4d7-661a-f59b-8cd399784e6e@delphij.net>
References:  <e79a8278-0fd8-532f-2a72-87d43cf27e7a@sentex.net> <56a4a35f-b4d7-661a-f59b-8cd399784e6e@delphij.net>

next in thread | previous in thread | raw e-mail | index | archive | help


> 23 apr. 2021 kl. 23:23 skrev Xin Li via freebsd-stable =
<freebsd-stable@freebsd.org>:
>=20
> On 4/23/21 13:53, mike tancsa wrote:
>> Starting to play around with RELENG_13 and wanted explore ZFS' built =
in
>> encryption.  Is there a best practices doc on how to do full disk
>> encryption anywhere thats not GELI based  ?  There are lots for=20
>> GELI,
>> but nothing I could find for native OpenZFS encryption on FreeBSD
>>=20
>> i.e box gets rebooted, enter in passphrase to allow it to boot kind =
of
>> thing from the boot loader prompt ?
>=20
> I think loader do not support the native OpenZFS encryption yet.
> However, you can encrypt non-essential datasets on a boot pool (that =
is,
> if com.datto:encryption is "active" AND the bootfs dataset is not
> encrypted, you can still boot from it).
>=20
> BTW instead of entering passphrase at loader prompt, if / is not
> encrypted, it's also possible to do something like
> =
https://lists.freebsd.org/pipermail/freebsd-security/2012-August/006547.ht=
ml
> .
>=20
> Personally I'd probably go with GELI (or other kind of full disk
> encryption) regardless if OpenZFS's native encryption is used because =
my
> primary goal is to be able to just throw away bad disks when they are
> removed from production [1].  If the pool is not fully encrypted, =
there
> is always a chance that the sensitive data have landed some =
unencrypted
> datasets and never gets fully overwritten.
>=20
> [1] Also keep in mind: https://xkcd.com/538/
>=20
> Cheers,
>=20
Yes, I=E2=80=99ve come to the same conclusion. This should be used on a =
data-zpool and not on the system-pool (zroot). Encryption is per =
dataset. Also if found that if the encrypted dataset is not mounted of =
some reason you will be writing to the parent unencrypted dataset.. At =
least it works for encrypted thumb_drive, i just posted this quick guide =
https://forums.freebsd.org/threads/freebsd-13-openzfs-encrypted-thumb-driv=
e.80008/ =
<https://forums.freebsd.org/threads/freebsd-13-openzfs-encrypted-thumb-dri=
ve.80008/>

/Peter=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4CFAA2E3-F8B0-41F3-BA2D-4802FC138E8C>