Beta v9.20.0 #7840
Draft
Beta v9.20.0 #7840
+1,214
−1,107
Conversation
This file contains hidden or 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
Release v9.19
We exceed the max 256 strategy matrix configurations
and `-nogui` compat => `--nogui` suggested to better use by `ocs-sr` code comment. No functional change.
and init DietPi v9.20 code version
The new dedicated CLI package from Debian, installed on RISC-V systems.
by default via device tree overlay. This makes it more transparent how to revert, comparing to alter the base device tree. And there is no overlay to set it to OTG mode (current DT default).
Bumps [actions/checkout](https://github.com/actions/checkout) from 5 to 6. - [Release notes](https://github.com/actions/checkout/releases) - [Changelog](https://github.com/actions/checkout/blob/main/CHANGELOG.md) - [Commits](actions/checkout@v5...v6) --- updated-dependencies: - dependency-name: actions/checkout dependency-version: '6' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
same as with did for Orange Pi 5
We enforced the downgrade only in case the Armbian repo was left in place, unintentionally one conditional too deep. Re-apply with DietPi v9.20 entirely unconditionally, aside of the impossibly high package version of course. But lower that one as well to >=15 to catch even possibly ancient cases. Latest Debian base-files is 14 until after Forky release in summer 2027. Even if then someone runs into this patch on Debian 15 Duke somehow, it will be just a noop, no harm.
* Additionally: Use MBR partition table for all RK356x SBCs. GPT is needed only for vendor U-Boot. It comes with some overhead and an additionally required reboot after partition resize (not possible to partprobe/partx a mounted filessystem on GPT partition), let's keep it simple where possible.
On Bullseye, gpg depends on libsqlite3-0, hence no need to install it.
Mount namespacing fails since systemd 259-rc1 with QEMU-emulation. This will cause a lot more services to fail, as common as sandboxing is, but doing a start with MariaDB. Also merging QEMU-emulation relaeted workarounds together.
and update repo URL which has been turned into a GitHub orga
Co-authored-by: MichaIng <[email protected]>
A numerical index is fragile, as it can change when drives are rescanned, as the whole array is rebuilt from scratch. A function was in place to align the numerical index with the mount target of the selected drive with a dedicated global variable, but it was not made use of in the `Unmount_Drive` function. This could cause false partitions or false whole drives to be formatted unintentionally, if a filesystem from that drive was mounted, and hence unmounted by the script. To harden the situation, the following changes have been done: * The mount source device node is used as index, using associative arrays, which is assured to never change and never require any manual alignment unless a drive is formatted differently. * The `Mount_Drive` and `Umount_Drive` functions themselves do not trigger a drive rescan anymore, instead this is done from the parent functions where needed. Notable during drive formatting, rescans could have happened multiple times if multiple filesystems from the formatted drive were mounted, unnecessary until the whole process finished, and the reason for the swapped drive index and false format target which triggered me to do this rework. * The format run function now uses a dedicated local variable to hold the device path to be formatted, to not rely on any external index. Unrelated changes: * Show drive/partition capacity for unmounted/unformatted drives, which can obtained via `lsblk`, and surely helps to identify certain drives or filesystems better than based on the device name only. Hide the UUID/device name based placeholder mount point for those, instead show unmounted/unformatted info where mouonted drives show the mount point. * Fix an error when selecting a mounted drive with a partition and select full drive format mode: It was attempted to additionally unmount the partition after it has been unmounted and formatted in a previous "all partitions" loop. Do a dedicated unmount attempt only if there is no partition table, but the filesystem was written on the bare drive without partition table. This is what it was meant for. * mergerfs mounts are now excluded from the mounted drive loop, since they are special-handled like tmpfs/ecryptfs/vboxsf/glusterfs mounts: their `/etc/fstab` entries are extracted and preserved untouched. * Do not skip the `blkid` cache when searching for unmounted filesystems. When executed as root, it is assured to check the current state, and despite that it performs faster when not using `/dev/null` as cache path. This has been change for the other calls already, but forgotten for this case. * Skip the sync call at the end of the drive detection loop, there is no real reason to do that * When doing a reboot, always exit the script immediately, since the reboot is done asynchronously and hence the script can continue unnecessarily before being killed by systemd * Some more variables used in info output and menu texts are shared between the format menu and the final format run function. * Shell code enhancements and format alignments
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.
No description provided.