public inbox for isar-users@googlegroups.com
 help / color / mirror / Atom feed
From: Henning Schild <henning.schild@siemens.com>
To: "[ext] Andreas J. Reichel" <andreas.reichel.ext@siemens.com>
Cc: <isar-users@googlegroups.com>
Subject: Re: [RFC v1 0/3] Fix additional apt repos with foreign keys
Date: Tue, 26 Feb 2019 16:06:59 +0100	[thread overview]
Message-ID: <20190226160659.33e7a6f8@md1za8fc.ad001.siemens.net> (raw)
In-Reply-To: <20190226134844.8632-1-andreas.reichel.ext@siemens.com>

Am Tue, 26 Feb 2019 14:48:41 +0100
schrieb "[ext] Andreas J. Reichel" <andreas.reichel.ext@siemens.com>:

> From: Andreas Reichel <andreas.reichel.ext@siemens.com>
> 
> This patch series fixes problems when adding a additional repos
> which need different gpg keys for authentication.
> 
> The patches are designed to make the existing 'API', i.e. bitbake
> variables work, not to solve the remaining design problems.
> 
> On basis of this series, we should discuss how to further proceed
> since there should be a destinction wether we change the bootstrapping
> apt source or if we change the apt source for additional packages.
> 
> If we change the bootstrapping apt source, we already need keys
> installed in the build environment to do the first debootstrap.
> 
> If we only want additional packages in the target rootfs, we only
> need to add keys inside the target chroot.
> 
> Currently this is not possible and requires additional bitbake
> variables, i.e. APT_KEYS_TARGET_PKGS, or APT_KEYS_TARGET_BOOTSTRAP.

I think it is useful to establish the trust twice and forget about
extra variables. People might want to bootstrap from "new/unknown"
mirrors, like i.e. the cache.
So use the variables we know to establish trust on the guy running
debootstrap and inside the chroots. That is two "apt-key" like in
your current patches.
But i would argue that you should play with "apt-key --keyring <file>".
The goal would be to create a keyring just for that one debootstrap
call, which you will remove/distrust later. For people not using docker
that will prevent "messing with the host". 

> Also the reason for the option I delete in patch 3 is unclear to me.
> This way we could never add additional repositories.

Good catch. That pattern is used in a few places, assuming that
isar-apt is the only repo that could have possibly changed. Maybe that
whole pattern should be revised and we go for plain "apt-get update"

Henning

> Andreas Reichel (3):
>   Fix path to user gpg-keys
>   Refactor gpg code to use apt code
>   Use all source lists in target root apt
> 
>  meta/classes/isar-bootstrap-helper.bbclass    | 14 ++++++++----
>  .../isar-bootstrap/isar-bootstrap.inc         | 22
> +++++++++---------- 2 files changed, 21 insertions(+), 15 deletions(-)
> 


      parent reply	other threads:[~2019-02-26 15:07 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-26 13:48 Andreas J. Reichel
2019-02-26 13:48 ` [RFC v1 1/3] Fix path to user gpg-keys Andreas J. Reichel
2019-02-26 15:33   ` Henning Schild
2019-02-26 13:48 ` [RFC v1 2/3] Refactor gpg code to use apt code Andreas J. Reichel
2019-02-26 14:07   ` cedric_hombourger
2019-02-26 15:57     ` Henning Schild
2019-02-26 15:39   ` Henning Schild
2019-02-26 13:48 ` [RFC v1 3/3] Use all source lists in target root apt Andreas J. Reichel
2019-02-26 15:06 ` Henning Schild [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190226160659.33e7a6f8@md1za8fc.ad001.siemens.net \
    --to=henning.schild@siemens.com \
    --cc=andreas.reichel.ext@siemens.com \
    --cc=isar-users@googlegroups.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox