:: Re: [DNG] unatternded upgrades by d…
Top Page
Delete this message
Reply to this message
Author: Olaf Meeuwissen
Date:  
To: KatolaZ
CC: dng
Subject: Re: [DNG] unatternded upgrades by default in Debian
Hi KatolaZ,

KatolaZ writes:

> On Wed, Feb 13, 2019 at 08:08:22AM +0900, Olaf Meeuwissen wrote:
>
> [cut]
>
>>
>> I'm using this on two Devuan machines at the office. Works fine for
>> me. Apart from running the `apt upgrade` it can also check for the need to
>> reboot and provides a number of knobs to fine tune when, how and under
>> what conditions to upgrade/reboot.
>>
>
> Hi Olaf,
>
> have you found unattended-upgrades automatically installed on those
> two boxen, or have you instead apt-get installed it afterwards?


Afterwards. Both are servers and I have a habit of doing

  sudo apt-mark $(apt-mark showmanual)
  # mark about two handful of packages I need as manual and then
  sudo apt purge -o APT::AutoRemove::RecommendsImportant=false \
                 -o APT::AutoRemove::SuggestsImportant=false


after the initial install ;-)
Of course, the /etc/apt/apt.conf on these machines have

APT::Install-Suggests "false"; // default behaviour
APt::Install-Recommends "false";

to keep things that way too.

>> Of course, you are free to invent a better wheel ;-)
>>
>> BTW, on Devuan I did run into an issue because of a missing dependency
>> that didn't happen on Debian courtesy of systemd. Can't seem to find
>> the bug report I filed way back when at the moment but it was rather
>> unceremoniously closed.
>
> If the bug you are referring to is thts one:
>
> https://bugs.devuan.org//cgi/bugreport.cgi?bug=78


No that's not the one. I filed it against Debian BTS.

> it was closed because we forked the package and solved the issue (or
> at least we thought so). We haven't had any other report of
> misfunctioning unattended-upgrades since. I think it looked like a
> regular bug, so at that time we must have concluded that no special
> "burial ceremony" was in order :D


Searching ... found it! Apologies for misremembering. The bug was with
systemd-sysv and happened on the one Debian server I'm also maintaining.
It was exposed by unattended-upgrades' `shutdown -r` in the absence of
dbus. That server also got the "zap everything that's not really needed
and keep it that way" APT treatment above ;-)

For those interested in the details, see

https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=898388

> And please, if you believe at any time that a bug report is still
> valid, and should have not been closed, just *reopen* it. There are
> humans over here, humans who don't have access to all your machines
> and cannot try all of the almost infinite number of combinations in
> which a package can find itself :)


Thanks for the reminder but I'm aware of that. It's just that some
humans are more amenable to reason than other (as that bugreport above
sort of shows ;-)

Hope this helps,
--
Olaf Meeuwissen, LPIC-2            FSF Associate Member since 2004-01-27
 GnuPG key: F84A2DD9/B3C0 2F47 EA19 64F4 9F13  F43E B8A4 A88A F84A 2DD9
 Support Free Software                        https://my.fsf.org/donate
 Join the Free Software Foundation              https://my.fsf.org/join