On 11/30/2018 08:16 PM, unman wrote:
On Fri, Nov 30, 2018 at 07:25:15PM -0500, Chris Laprise wrote:
On 11/30/2018 06:06 PM, Marek Marczykowski-Górecki wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
[moved discussion from ticket #2065]
On Fri, Nov 30, 2018 at 11:44:21AM +0000, Patrick Schleizer wrote:
Is it possible to make changes in Qubes packages only affecting Debian releases
equal or greater than 9?
Inability to do so would make development difficult. Not a secret that I am not
a big fan of carrying around oldstable versions support. Lot effort for little
to zero popularity? I'd not release new Qubes packages for debian-8 in order to
not break them. Only debian-9 and above. Otherwise too many targets required to
test / too much development effort, imo.
Actually the same problem applies to Fedora, with an exception that many
of those templates are not even supported upstream anymore (in contrast
to oldstable, which is still part of Debian LTS, until 2020).
So here are actually two questions:
1. Do we want to stop publishing new packages for templates not
supported upstream anymore?
2. Do we want to stop publishing new packages for old templates still
supported upstream?
I think the answer for the first one should be "YES". Look at the insane
count of Fedora versions for R3.2 on supported-versions[1].
We just need a process for doing that:
- an announcement? we had already some[2]
- some method for marking this on supported-versions page (simply
remove version? or should we retain what were initially supported?)
In fact, I've already went ahead and disabled building packages for old
Fedora versions[3], mostly to reduce build time. Oh, Fedora 27 upstream
support ends today, so we could drop this one from the list too. And
post a reminder to migrate to Fedora 28 (or have we already done it? I
can't find it).
Now, about the second question. It is about Debian 8 and Whonix 13
(based on Debian 8). Since Debian 8 was included originally
in both Qubes 3.2 and 4.0, many people surely do use it. I don't have
hard numbers for that (in theory we could collect such stats from our
updates server, but we don't do that right now). Since Debian 8 still
has upstream security support, there is no security-based reason for
dropping it. But as Patrick said, it makes it difficult to maintain.
Especially since Debian 8 is much older than oldest supported Fedora...
I think Debian oldstable templates should be supported through the support
lifespan that the distro offers. In general, ending support prematurely is
destabilizing; PC users shouldn't be forced into rapid-change environments
if it can be avoided.
Long term support is one of the main advantages of offering Debian. (The
other two main advantages over Fedora are larger software library and
fully-signed repositories.)
Conversely, one of the things that makes using Fedora so annoying is the
relentless (often trivial and buggy) updates for about 2/3 of its 'always
young' lifecycle, then having to do OS upgrades after only 12-15 months of
use.
With that said, I don't think long-term Debian support should necessarily
apply to Whonix, which is its own distro in a sense.
Completely agree with Chris. We should maintain oldstable throughout
LTS for Debian.
Whonix 13 already lacks suport from WhonixQubes, so perhaps it should
be dropped from Qubes support also - although as there appear to be
a number of issues affecting the update to 14 it might be best to see
those resolved first before dropping Qubes support.
unman
FWIW, I was surprised when R4.0 was released with Debian 8 (with only 3
months until EOL) instead of 9.
Although I'd prefer that Debian be supported past oldstable through its
LTS phase, I understand what a burden that is to a small project like Qubes.
And in the specific Debian 8 case, an announcement about EOL was already
made. IMHO it was also a temperamental version that encompassed a
problematic era in Linux userspace, so if it were my decision and I
wanted to start supporting Debian LTS to the five-year mark I wouldn't
start that policy until version 9.
--
Chris Laprise, tas...@posteo.net
https://github.com/tasket
https://twitter.com/ttaskett
PGP: BEE2 20C5 356E 764A 73EB 4AB3 1DC4 D106 F07F 1886
--
You received this message because you are subscribed to the Google Groups
"qubes-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to qubes-devel+unsubscr...@googlegroups.com.
To post to this group, send email to qubes-devel@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/qubes-devel/b192add3-f12f-f075-b6a8-82b7ff07f8d7%40posteo.net.
For more options, visit https://groups.google.com/d/optout.
