From: Henning Schild <henning.schild@siemens.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: "Q. Gylstorff" <Quirin.Gylstorff@siemens.com>,
<isar-users@googlegroups.com>,
"Su, Bao Cheng (RC-CN DF FA R&D)" <baocheng.su@siemens.com>
Subject: Re: [PATCH] classes/image-postproc-extension: Remove /etc/machine-id
Date: Wed, 20 Apr 2022 11:00:02 +0200 [thread overview]
Message-ID: <20220420110002.63a60c14@md1za8fc.ad001.siemens.net> (raw)
In-Reply-To: <f56c08a8-e9ca-585b-78b4-be5ce6b9bc9d@siemens.com>
Am Tue, 19 Apr 2022 18:15:10 +0200
schrieb Jan Kiszka <jan.kiszka@siemens.com>:
> On 19.04.22 17:36, Q. Gylstorff wrote:
> > From: kas <kas@example.com>
> >
> > In a read-only system the machine id should be deleted or
> > set to `unitialized\n`[1].
> >
> > Systemd will generate a new machine-id during the first boot.
> > In the case of a read-only root file system Systemd generates a
> > mount point with the machine id. If an overlay for /etc is used
> > this creates a mount conflict. To avoid the conflict between the
> > overlay filesystem and systemd /etc/machine-id mount point deleted
> > the file /etc/machine-id.
> >
> > [1]: https://systemd.io/BUILDING_IMAGES/
> >
> > Signed-off-by: Quirin Gylstorff <quirin.gylstorff@siemens.com>
> > ---
> > meta/classes/image-postproc-extension.bbclass | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/meta/classes/image-postproc-extension.bbclass
> > b/meta/classes/image-postproc-extension.bbclass index
> > ca520273..0c412c0d 100644 ---
> > a/meta/classes/image-postproc-extension.bbclass +++
> > b/meta/classes/image-postproc-extension.bbclass @@ -57,7 +57,7 @@
> > ROOTFS_POSTPROCESS_COMMAND =+ "image_postprocess_machine_id"
> > image_postprocess_machine_id() { # systemd(1) takes care of
> > recreating the machine-id on first boot sudo rm -f
> > '${IMAGE_ROOTFS}/var/lib/dbus/machine-id'
> > - sudo install -m 644 '/dev/null'
> > '${IMAGE_ROOTFS}/etc/machine-id'
> > + sudo rm -f '${IMAGE_ROOTFS}/etc/machine-id'
> > }
> >
> > ROOTFS_POSTPROCESS_COMMAND =+ "image_postprocess_sshd_key_regen"
>
> This partially reverts 98d6a72d3064. Did you read that commit and can
> you comment on it in your patch?
In fact the man pages changes a bit from buster to bullseye, but having
the machine-id being an empty file still seems the most generic way
which should cover both older and newer systemd. However we will never
have a "first boot" "ConditionFirstBoot=yes" when using that empty
file, i tend to think removal is the correct way ... at least for newer
systemd.
https://manpages.debian.org/buster/systemd/machine-id.5.en.html
says
> Having an empty file in place is useful because it allows a temporary
> file to be bind-mounted over the real file, in case the image is used
> read-only.
https://manpages.debian.org/bullseye/systemd/machine-id.5.en.html
says
> 3.If /etc/machine-id exists and is empty, a boot is not considered
> the first boot. systemd will still bind-mount a file containing the
> actual machine-id over it and later try to commit it to disk (if
> /etc/ is writable).
Which makes me wonder if we want systemd to know that it is the first
boot, i bet we would want that ... and not want an empty machine id.
Also the link quirin sent ... coming directly from systemd says one
should remove the file.
Maybe things really changed between buster and bullseye and we need to
do other things for more recent systemd.
Henning
> Jan
>
next prev parent reply other threads:[~2022-04-20 9:00 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-04-19 15:36 Q. Gylstorff
2022-04-19 16:15 ` Jan Kiszka
2022-04-20 9:00 ` Henning Schild [this message]
2022-04-19 19:10 ` Anton Mikanovich
2022-04-20 7:28 ` [PATCH v2] " Quirin Gylstorff
2022-04-20 13:00 ` [PATCH v3] " Quirin Gylstorff
2022-05-14 14:04 ` Anton Mikanovich
2022-04-20 8:44 ` [PATCH] " Henning Schild
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=20220420110002.63a60c14@md1za8fc.ad001.siemens.net \
--to=henning.schild@siemens.com \
--cc=Quirin.Gylstorff@siemens.com \
--cc=baocheng.su@siemens.com \
--cc=isar-users@googlegroups.com \
--cc=jan.kiszka@siemens.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