Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 28 Apr 2020 07:25:35 +0100
From:      Steve O'Hara-Smith <>
To:        Aryeh Friedman <>
Cc:        Daniel Feenberg <>, Arne Steinkamm <>, Ihor Antonov <>, FreeBSD Mailing List <>
Subject:   Re: Wayland on FreeBSD
Message-ID:  <>
In-Reply-To: <>
References:  <> <> <5058973.kMyvyFPq5o@amos> <> <> <> <> <> <> <> <> <>

Next in thread | Previous in thread | Raw E-Mail | Index | Archive | Help
On Tue, 28 Apr 2020 01:56:57 -0400
Aryeh Friedman <> wrote:

> On Tue, Apr 28, 2020 at 1:47 AM Steve O'Hara-Smith <>
> wrote:
> > On Mon, 27 Apr 2020 21:07:08 -0400
> > Aryeh Friedman <> wrote:
> >
> > > The jury seems to still out (at best) on the question of block vs.
> > > file. I suspect the main issue are you NAS (file storage) or SAN
> > > (block storage):
> >
> >         Both have their uses and both are used heavily in large scale
> > storage (tens of petabytes upwards). As always requirements dictate the
> > best solution.
> >
> >
> The point with OpenStack is the choice should of been left to the
> user/admin instead they just out of the blue decided everything must be

	Yep such choices should always be left to the admin as long as they
are real choices, which AFAICS this is.

> the entire point of this subthread was to show that OpenStack picked wrong
> and did so based on standard recommendations in the Linux community at the
> time (2013-2014) and that those recommendations where/are still being made
> and are demonstrably wrong.

	Following recommendations without understanding the reasons and
consequences is not good engineering practice. Fortunately Linux (the
kernel) and every Linux distribution I know of does a decent job of
supporting NFS and even does as well as possible supporting CIFS (SMB)
so both block and file based network storage are supported.

	There may be a vocal presence saying block is good and NFS is a
disaster area in the Linux community (of course SMB/CIFS is intolerable too
because it comes from Microsoft - who it hard to interoperate with). They
can be safely ignored IMHO.

> The other main point was to show why source
> distrubtion is better then binary distrubtion (if OpenStack was available
> in easy to compile source form [which is and was not] it would of been
> trivial to do a after market patch to not require iSCSCI)


Steve O'Hara-Smith                          |   Directable Mirror Arrays
C:\>WIN                                     | A better way to focus the sun
The computer obeys and wins.                |    licences available see
You lose and Bill collects.                 |

Want to link to this message? Use this URL: <>