-
Tue Jan 20 2026 EL Errata <el-errata_ww@oracle.com> - 239-82.0.8.el8_10.13
- Return to newest device wins for competing symlink claims [Orabug: 38718322]
- Fix collection of device units with conflicting devlinks [Orabug: 38599776]
- Re-apply fix for devlink device units on lvm rename [Orabug: 38491067]
- Removed fix for devlink device units on lvm rename [Orabug: 38659832]
- Fix leak of old devlink device units on lvm rename [Orabug: 38491067]
- Stash the dbus subscriber list when we disconnect from the bus [Orabug: 38028720]
- Drop systemd-nspawn delay on failing to reset loginuid [Orabug: 37782633]
- coredump: use %d in kernel core pattern - CVE-2025-4598
- Fixes podman quadlet doesn't work in rootless mode [Orabug: 36076771]
- Drastically simplify caching of cgroups members mask
- drop IN_ATTRIB from parent directory inotify watches [Orabug: 36780432]
- Udevd: add an extra configurable timeout before udevd kills workers [Orabug: 36424686]
- Fixed deletion issue for symlink when device is opened [Orabug: 36228608]
- Fix local-fs and remote-fs targets during system boot (replaces old Orabug: 25897792) [Orabug: 35871376]
- 1A) Add "systemd-fstab-generator-reload-targets.service" file [Orabug: 35871376]
- 1B) Add required rpms for correct kickstart/systemd functionality within systemd.spec [Orabug: 35871376]
- 1C) Important: Review 1902-systemd-fstab-generator-reload-targets.patch for important build details/steps [Orabug: 35871376]
- Prevent duplicate uuid device to replace existing one in udev [Orabug: 35987487]
- Backport upstream pstore dmesg fix [Orabug: 34850699]
- mount: flush out cycle state on DEAD->MOUNTED only, not the other way round [Orabug: 35454661]
- core/mount: adjust deserialized state based on /proc/self/mountinfo [Orabug: 35454661]
- Prevent duplicate label to replace exsisting one in udev [Orabug: 34898273]
- Oracle-Redhat Errata ELSA-2023:3837 CVE-2023-26604 OLERRATA-43629
- Detect podman as separate container type [Orabug: 31922204]
- improve container detection logic [Orabug: 31922204]
- mount: flush out cycle state on DEAD->MOUNTED only, not the other way round [Orabug: 35454661]
- core/mount: adjust deserialized state based on /proc/self/mountinfo [Orabug: 35454661]
- Prevent duplicate label to replace existing one in udev [Orabug: 34898273]
- Standardize ioctl (BTRFS_IOC_QGROUP_CREATE) check and return -ENOTCONN, if quota is not enabled [Orabug: 34694253]
- Disable unprivileged BPF by default [Orabug: 32870980]
- udev rules: fix memory hot add and remove [Orabug: 31310273]
- fix to enable systemd-pstore.service [Orabug: 30951066]
- journal: change support URL shown in the catalog entries [Orabug: 30853009]
- set "RemoveIPC=no" in logind.conf as default for OL7.2 [Orabug: 22224874]
- allow dm remove ioctl to co-operate with UEK3 (Vaughan Cao) [Orabug: 18467469]
- Backport upstream patches for the new systemd-pstore tool (Eric DeVolder) [OraBug: 30230056]
- Removed unneeded patches (Already provided upstream or not required)
- 1902-Fix-missing-netdev-for-iscsi-entry-in-fstab.patch [Orabug: 25897792]
- 2002-orabug31420486-pstore-introduce-tmpfiles.d-systemd-pstore.conf.patch [Orabug: 31420486]
- 2009-login-add-a-missing-error-check-for-session_set_lead.patch (#2158167)
- 2010-logind-reset-session-leader-if-we-know-for-a-fact-th.patch (#2158167)
- 2011-sulogin-fix-control-lost-of-the-current-terminal-whe.patch (#2227769)
- systemd.spec: prevent 'myhostname' from being appended on upgrade (#2187761) (#2227769)
- Updated mod_nss() and readlink /etc/nsswitch.conf sections (#2187761)
- systemd.spec: mod_nss() and readlink /etc/nsswitch.conf sections (#2187761)
-
Mon Dec 08 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.13
- logind: fix crash in logind on user-specified message string (RHEL-132317)
-
Fri Dec 05 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.12
- Revert "run: update checks to allow running with a user's bus" (RHEL-118835)
-
Tue Dec 02 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.11
- run: update checks to allow running with a user's bus (RHEL-118835)
-
Tue Dec 02 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.10
- hwdb: add ACCEL_LOCATION property to parse_hwdb.py (RHEL-130979)
- hwdb: update ACCEL_LOCATION property to use Or instead of QuotedString (RHEL-130979)
- test: support general properties in hwdb files (RHEL-130979)
- hwdb: Relax parsing script to allow 0 and 1 for all ID_* properties (RHEL-130979)
- hwdb: allow spaces in usb: matches and similar patterns (RHEL-130979)
- test: fix parsing of 60-seat.hwdb and 60-keyboard.hwdb (RHEL-130979)
- parse_hwdb: fix compatibility with pyparsing 2.4.* (RHEL-130979)
- login: use parse_uid() when unmounting user runtime directory (RHEL-132175)
- pid1: do not use generated strings as format strings (#19098) (RHEL-132317)
- core/transaction: make merge_unit_ids() always return NUL-terminated string (RHEL-132317)
- core/transaction: make merge_unit_ids() return non-NULL on success (RHEL-132317)
- core/transaction: do not log "(null)" (RHEL-132317)
-
Wed Nov 05 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.9
- cryptsetup-generator: refactor add_crypttab_devices() (RHEL-38859)
- cryptsetup-generator: continue parsing after error (RHEL-38859)
-
Thu Oct 02 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.8
- test-execute: let's ignore the difference between CLD_KILLED and CLD_DUMPED (RHEL-108744)
- test-execute: turn off coredump generation in test services (RHEL-108744)
- test: introduce TEST-53-TIMER (RHEL-108744)
- test: restarting elapsed timer shouldn't trigger the corresponding service (RHEL-108744)
- test: check the next elapse timer timestamp after deserialization (RHEL-108744)
- timer: don't run service immediately after restart of a timer (RHEL-108744)
- Revert "test: extend testcase to ensure controller membership doesn't regress" (RHEL-9322)
-
Fri Aug 29 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.7
- cgroup-util: add mask definitions for sets of controllers supported by cgroupsv1 vs. cgroupsv2 (RHEL-9322)
- cgroup: dump delegation mask too (RHEL-9322)
- cgroup: units that aren't loaded properly should not result in cgroup controllers being pulled in (RHEL-9322)
- cgroup: be more careful with which controllers we can enable/disable on a cgroup (RHEL-9322)
- cgroup: extend reasons when we realize the enable mask (RHEL-9322)
- cgroup: drastically simplify caching of cgroups members mask (RHEL-9322)
- cgroup: when we unload a unit, also update all its parent's members mask (RHEL-9322)
- test: extend testcase to ensure controller membership doesn't regress (RHEL-9322)
-
Mon Aug 25 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.6
- dbus: stash the subscriber list when we disconenct from the bus (RHEL-75081)
- manager: s/deserialized_subscribed/subscribed_as_strv (RHEL-75081)
- bus-util: do not reset the count returned by sd_bus_track_count_name() (RHEL-75081)
- core: do not disconnect from bus when failed to install signal match (RHEL-75081)
- sd-bus/bus-track: use install_callback in sd_bus_track_add_name() (RHEL-75081)
- core/manager: restore bus track deserialization cleanup in manager_reload() (RHEL-75081)
- core/manager: drop duplicate bus track deserialization (RHEL-75081)
-
Wed Mar 05 2025 systemd maintenance team <systemd-maint@redhat.com> - 239-82.5
- man: be even clearer that tmpfiles user/group/mode are applied on existing inodes (RHEL-77145)
- Revert "man: fix description of --force in halt(8) (#7392)" (RHEL-81056)
- man: explicitly document that "reboot -f" is different from "systemctl reboot -f" (RHEL-81056)