Rebase on upstream most recent version 0.22.0-5 (93cae9c88e964d6485ad8314ae65deb0ab676862)#148
Conversation
This commit addresses the issue where the `/boot/grub2/grub.cfg` file is overwritten during the upgrade process by an old RHEL7 configuration leftover on the system, causing the system to fail to boot. The problem occurs on hybrid Azure images, which support both UEFI and Legacy systems and have both `grub-pc` and `grub-efi` packages installed. It is caused by one of the scriplets in `grub-efi` which overwrites the old configuration. If old configuration is detected, this actor regenerates the grub configuration using `grub2-mkconfig -o /boot/grub2/grub.cfg` after installing rpms to ensure the correct boot configuration is in place. The fix is applied specifically to Azure hybrid cloud systems. JIRA: RHEL-38255 (cherry picked from commit 9966eb1)
Previosly detection of Azure hybrid image was tightly coupled with process of converting grubenv symlink to a regular file. Since there exists other issues relating to hybrid images it is worth to separate these two concepts. This commit modifies the ScanHybridImage actor so that it produces a message whel WALinuxAgent is detected or we are booted in bios and ESP partition is mounted and we are running on Hyper-V (sign of a hybrid image). New CheckGrubenvToFile actor is responsible for detection of grubenv symlink on hybrid images and tasks ConvertGrubenvToFile that is later responsible for the actual conversion. (cherry picked from commit 1964c69)
Instead of the leapp framework contribution guidelines. Also update the leapp-repository contribution guidelines to point to the new place of the leapp framework Python coding guidelines (moved under oamg/leapp@123f270). (cherry picked from commit ba074a4)
The DNF configuration has not been loaded when trying to get information about available module streams (library module.py). This causes a traceback e.g. on systems which must access DNF repositories via a proxy. This patch introduces loading the DNF configuration before trying to access remote resources. Jira: RHEL-39095 (cherry picked from commit be6d232)
This commit is setting the localpkg_gpgcheck DNF option within the rhel-upgrade plugin to 0 (disabled). The upgrade process has been blocked with "Error: GPG check FAILED" when dnf has been configured to apply gpg check also on local packages (localpkg_gpgcheck=1). That's because the bundled leapp*-deps meta packages, which are managing Leapp and leapp-repository dependencies during the transition to the new system, are not signed by the Red Hat key. Therefore, this option needs to be disabled. Jira: RHEL-47472 (cherry picked from commit a4d10ad)
- Disable IPU for: AlmaLinux 9.5 -> 10.0 CentOS 7 -> 8, 7.9 -> 8 All Scientific Linux IPUs - Allow IPU: Rocky only 8.10 -> 9.6 - Drop from the list of supported versions: AlmaLinux, Rocky versions 8.6, 8.7, 8.8, 8.9 Oracle 8 all versions Almalinux 9.4, 9.5 Rocky 9 all versions
(cherry picked from commit acaed7e)
(cherry picked from commit b9d7e96)
(cherry picked from commit 803aebd)
Introduce a new mysql_check actor which checks for presence of the mysql_server package and reports related recommendations for upgrade. Jira: RHEL-5459 (cherry picked from commit 7c5d7f7)
Currently only the format of a version specified by LEAPP_DEVEL_TARGET_RELEASE is checked, when specified via --target cmdline argument it isn't, which is a bug. This patch fixes the bug, by enabling the check for --target too. NOTE: these are only format checks, the state of support is already checked later in the upgrade process by the checktargetversion actor. Jira: RHEL-96238 (cherry picked from commit ce71a44)
Removing el7toel8 repository since the upgrade path is no longer supported. Changes: * el7toel8 repo removed * Reflect this change in the documentation * Makefile adjusted * Adjusted tests for scancustommodifications actor, which tested actors from el7toel8 repo. Jira: RHELMISC-11005 (cherry picked from commit b6e84f7)
(cherry picked from commit b8d942a)
Ignore source/debug repositories when running dnf repolist to check what repositories are available for the upgrade. Should prevent needless crashes when the source/debug repositories are malformed, e.g., when the maintainer forgot to sync their content. (cherry picked from commit 7f9df48)
The current implementation of the checkrhui actor and its modification of rhui-associated target userspace preparation instructions contains too coarse conditions on what should happen on specific upgrade paths. In other words, changes originally intended only for 8>9 are applied also during 9>10. This patch fixes this issue -- when performing 9>10 on AWS, we do not bootstrap the RHUI client, as it has too many dependencies. Instead, the leapp-rhui-aws package contains all repositories, certs and keys that grant repository access. Jira-ref: RHEL-64910 (cherry picked from commit 2471460)
Co-Authored-By: jinkangkang.jkk <1547182170@qq.com> (cherry picked from commit d0f6c44)
The RHEL10's Azure RHUI client is signed using a different GPG key than RHEL9 client. Therefore, we need to have the key available when we try to bootstrap the RHEL10 client during the upgrade. Jira-ref: RHEL-64911 (cherry picked from commit 08196d4)
…cloud (cherry picked from commit 7468b10)
* Inhibit upgrade on systems with cgroupsv1 on 9to10 cgroups-v1 were deprecated in RHEL 9 and are unsupported in RHEL 10. The option to switch to cgroups-v1 is ignored. Users have to migrate to cgroups-v2 in order to upgrade. Both legacy and hybrid modes are unsupported. Only unified hierarchy (cgroups-v2) is supported. More info on how the modes are switched is at: https://www.freedesktop.org/software/systemd/man/247/systemd.html#systemd.unified_cgroup_hierarchy and https://www.freedesktop.org/software/systemd/man/247/systemd.html#systemd.legacy_systemd_cgroup_controller. Jira: RHEL-81212 (cherry picked from commit 3356c04)
Update the PES data to the latest fresh content. (cherry picked from commit 93cae9c)
Thank you for contributing to the Leapp project!Please note that every PR needs to comply with the Leapp Guidelines and must pass all tests in order to be mergeable.
Packit will automatically schedule regression tests for this PR's build and latest upstream leapp build.
Note that first time contributors cannot run tests automatically - they need to be started by a reviewer. It is possible to schedule specific on-demand tests as well. Currently 2 test sets are supported,
See other labels for particular jobs defined in the Please open ticket in case you experience technical problem with the CI. (RH internal only) Note: In case there are problems with tests not being triggered automatically on new PR/commit or pending for a long time, please contact leapp-infra. |
No description provided.