bring filesystem package inline with Fedora (i.e. don't include files in the filesystem package) #8311
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ddstreetmicrosoft
force-pushed
the
ddstreet/filesystem
branch
2 times, most recently
from
15e3719 to
d0dcfa4
Compare
anphel31
reviewed
Mar 8, 2024
PawelWMS
reviewed
Mar 8, 2024
Comment on lines
+2
to
+5
| # Default to no check, because it pulls in other packages at build, | ||
| # and that will cause "circular dependency" problems for the Azure | ||
| # Linux toolkit | ||
| %bcond check 0 |
There was a problem hiding this comment.
Why do we need to hard-code a disabled test? We shouldn't implement a separate mechanism to controlling tests runs over what we already try to do with with_check. What happens when you use with_check instead and remove this line? The circular dependencies sound worrying in this case.
There was a problem hiding this comment.
reduce delta from upstream, and i'd like to reenable it when/if we are not so restricted by "circular dependencies"
There was a problem hiding this comment.
I missed a BR for tcsh - that would be an issue for toolchain builds, I see now what you mean.
I think it would still be worth to enable the tests to run during non-toolchain builds, so we keep setup tested. How that works, is all the BRs behind 0%{with_check} will be ignored during the non-test build and the test build of the package will wait until its dependencies are present (BRs behind 0%{with_check} never cause circular dependencies). The %check failures during the toolchain build are ignored, so they shouldn't fail the build as well. Once we get testing sorted out during the toolchain phase (or we restructure the build process in another way), then the circular dependencies issue may come back, but at that point we'll need to solve it globally for all packages anyway.
ddstreetmicrosoft
force-pushed
the
ddstreet/filesystem
branch
from
d0dcfa4 to
600ab53
Compare
jslobodzian
requested changes
Mar 8, 2024
PawelWMS
reviewed
Mar 11, 2024
PawelWMS
reviewed
Mar 11, 2024
PawelWMS
reviewed
Mar 11, 2024
Also drop unnecessary modprobe.d/usb.conf file.
…m package Also remove the old signature file from extended, and set the setup package to require filesystem package.
ddstreetmicrosoft
force-pushed
the
ddstreet/filesystem
branch
from
600ab53 to
9a68280
Compare
ddstreetmicrosoft
force-pushed
the
ddstreet/filesystem
branch
from
9a68280 to
c96f6ce
Compare
just waiting on your go ahead to merge. |
| @@ -1186,6 +1188,9 @@ rm -f %{name}.lang | |||
| # %autochangelog. So we need to continue manually maintaining the | |||
| # changelog here. | |||
| %changelog | |||
| * Wed Mar 06 2024 Dan Streetman <[email protected]> - 255-8 | |||
| - move /var/log/* files from filesystem to systemd package | |||
There was a problem hiding this comment.
Just double-checking if any files need to be provided by the systemd package, since there isn't a specific /var/log/* diff here (other than the filesystem conflicts which deprecate specific files under /var/log that filesystem used to provide).
There was a problem hiding this comment.
right, they are actually already being provided by the current (fedora-based) systemd package, so our current situation has them being provided by both filesystem and systemd packages:
ddstreet [ ~ ]$ rpm -q filesystem
filesystem-1.1-19.azl3.x86_64
ddstreet [ ~ ]$ rpm -qlv filesystem|grep var/log
drwxr-xr-x 2 root root 0 Mar 8 09:18 /var/log
-rw------- 1 root root 0 Mar 8 09:18 /var/log/btmp
-rw-rw-r-- 1 root utmp 0 Mar 8 09:18 /var/log/lastlog
-rw-r--r-- 1 root root 0 Mar 8 09:18 /var/log/wtmp
ddstreet [ ~ ]$ rpm -q systemd
systemd-255-7.azl3.x86_64
ddstreet [ ~ ]$ rpm -qlv systemd|grep var/log
-rw-rw---- 1 root root 0 Mar 8 16:50 /var/log/btmp
drwxr-xr-x 2 root root 0 Mar 8 16:50 /var/log/journal
-rw-rw-r-- 1 root root 0 Mar 8 16:50 /var/log/lastlog
drwx------ 2 root root 0 Mar 8 16:50 /var/log/private
-rw-rw-r-- 1 root root 0 Mar 8 16:50 /var/log/wtmp
There was a problem hiding this comment.
Any chance containers will be writing to these folders and have issues? We don't have systemd in our containers by default
There was a problem hiding this comment.
i don't see why it would cause a problem?
ddstreetmicrosoft
force-pushed
the
ddstreet/filesystem
branch
from
c96f6ce to
af48562
Compare
|
And here is the diff between our current filesystem scripts and the updated filesystem scripts: |
Note that our mis-assignment of
|
|
and files diff (with each listing edited to remove differing sizes and timestamps): |
Note that we need to fix our |
dmcilvaney
reviewed
Mar 13, 2024
| # Configuration files | ||
| # | ||
| cat > %{buildroot}/etc/passwd <<- "EOF" | ||
| root:x:0:0:root:/root:/bin/bash |
There was a problem hiding this comment.
Is setup creating these users now?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Move the files contained in the
filesystempackage out into the appropriate separate packages (mostly thesetuppackage), and update thefilesystempackage to use Fedora packaging.Fixes: #6927