The Discoverable Partitions Specification (DPS)
is designed for appliance-like and image-based devices: one OS on a GPT disk where every partition
carries a type GUID that tells its role (root, /usr, /home, /srv, /var, /var/tmp, swap, ESP,
...). With that, systemd can find and mount (and unlock, if LUKS) the partitions without fstab,
crypttab or root=. Auto-discovery is only a fallback: it is ignored for anything listed in
fstab, and it only looks at the disk the system booted from.
Automate the process of fixing CVEs (Common Vulnerabilities and Exposures) in the Agama project by:
- Discovering new CVEs assigned to Agama in Bugzilla
- Classifying CVEs by affected branch and component
- Fixing CVEs by updating dependencies in the appropriate branches
- Documenting fixes with proper changelog entries including Bugzilla references
Implement a GitHub Actions workflow that periodically scans for CVEs affecting Agama dependencies across multiple branches (master + maintenance branches like SLE-16, SLE-16.1, etc.), cross-references them with SUSE bugzilla reports, and automatically creates GitHub issues for tracking.
We are intentionally letting the SUSE security team drive the CVE tracking process rather than proactively looking for all CVEs ourselves. This design decision is based on the following practical considerations:
Based on feedback collected by AI, see https://github.com/jreidinger/agama-feedback-reports/tree/main/reports:
- Lack of cancel/abort
- Allow to quickly navigate between sections (storage, network, etc) without going back to overview.
- Install individual packages
- Missing wifi with online iso -> ends with a software issue
- solved: agama-project/agama#3768
How it works:
- YaST packages use semantic-like versioning:
MAJOR.MINOR.PATCH.MAINTENANCE - Example:
5.0.3.1→ version 5.0.3 with maintenance update 1 - The fourth number was recently added for maintenance releases
Critical limitation: The major and minor versions are completely frozen and meaningless.
Analysis of crates compiled in multiple versions across the workspace.
Run with: cargo tree --duplicates
TL;DR — All 13 conflicts are purely transitive. The workspace itself uses modern versions throughout. No workspace Cargo.toml changes can fix these; they require upstream crate updates.
What you are experiencing is the classic behavior of a standard, non-clustered filesystem (like ext4, Btrfs, or XFS) on shared raw block storage. Standard filesystems assume they have exclusive control over the drive—they cache disk structures in RAM, so when Machine B modifies or deletes partitions on the disk, Machine A's kernel has no idea the disk structure changed and corrupts its memory cache or throws read errors.
To stop two machines from accessing the same LUN at the same time, you have two primary ways to fix this: Target-Side Access Control (The Correct Fix) or Initiator/Cluster-Side Locking.
The safest and most reliable place to prevent concurrent access is at the iSCSI Target (the storage server or SAN exporting the LUN).
Date: 2026-08-03. All statements below were checked against the current upstream sources (libstorage-ng master, lsscsi master, util-linux master, torvalds/linux master, systemd main, sg3_utils master), against the local system (openSUSE, lsscsi 0.32) and against the local checkouts of yast2-storage-ng and libstorage-ng.
- There is no authoritative list of "data transports" for Linux. Neither the kernel nor any
See gist https://gist.github.com/ancorgs/07d8c772226cfd1794631f284b5ad903, section "Searching with drives by their partitions".
I am trying to evaluate the JSON schema proposed in the gist to search for partitions. There are several details I am not convinced about.
max: 0 is used to indicate that none of the partitions matches the given condition: