Affected versions: SLES 16

📖 ~4 min read  •  Source: SUSE advisory RHSA-2026:1902 (see also SUSE bugzilla)

Related CVEs: CVE-2026-24049 CVE-2022-40898

Upstream summary: wheel is a command line tool for manipulating Python wheel files, as defined in PEP 427. In versions 0.40.0 through 0.46.1, the unpack function is vulnerable to file permission modification through mishandling of file permissions after extraction. The logic blindly trusts the filename from the archive header for the chmod operation, even though the extraction process itself might have sanitized the path. Attackers can craft a malicious wheel file that, when unpacked, changes

Table of contents
  1. Symptom & Impact
  2. Environment & Reproduction
  3. Root Cause Analysis
  4. Quick Triage
  5. Step-by-Step Diagnosis
  6. Solution – Primary Fix
  7. Solution – Alternative Approaches
  8. Verification & Acceptance Criteria
  9. Rollback Plan
  10. Prevention & Hardening
  11. Related Errors & Cross-Refs
  12. References & Further Reading

Symptom & Impact

python313-wheel on SLES 16 runs only when something invokes it, so there is no crash loop and no failed unit to notice — the exposure is input-driven. A developer build, a CI job, a package rebuild or triage tooling that runs objdump, readelf or gdb across an untrusted binary can hit a malformed section header and corrupt memory inside the toolchain process, running code as the build user. On a shared builder or CI runner that means tampered build output rather than downtime. python313-wheel is a build-time developer tool that exits when the command finishes, so there is no unit to restart — only in-flight builds and long-lived helpers such as distcc or clangd still hold the old binaries.

Environment & Reproduction

Reproduction targets SLES 16. Confirm release, registration, and installed package:

cat /etc/os-release
SUSEConnect --status-text
SUSEConnect --list-extensions 2>/dev/null | head -30
rpm -q python313-wheel
zypper info python313-wheel | head -20

Exercise the workload that uses python313-wheel while collecting:

rpm -qa --qf '%{VENDOR}\t%{BUILDHOST}\t%{BUILDTIME:date}\t%{NAME}-%{VERSION}-%{RELEASE}\n' | awk -F'\t' '$1 !~ /SUSE/'  # packages NOT built by SUSE are your own builds - these were compiled with the vulnerable python313-wheel and need rebuilding
readelf -p .comment /usr/local/bin/YOURAPP  # prints e.g. 'GCC: (SUSE Linux) 7.5.0' - the compiler that actually built it (binutils; empty output just means the section was stripped)
objdump -s -j .comment /usr/local/bin/YOURAPP  # the same evidence via objdump (binutils)
sudo journalctl -xe --no-pager | tail -200
sudo tail -200 /var/log/zypp/history
sudo tail -200 /var/log/audit/audit.log
# For SUSE support, bundle evidence with supportconfig:
sudo supportconfig -R /var/tmp -B python313-wheel

Root Cause Analysis

Root cause is documented in SUSE advisory RHSA-2026:1902. SUSE security maintainers shipped fixes in the corresponding python313-wheel update for SLES 16; running an outdated build leaves the host exposed to the failure modes described in the advisory. Correlate zypper history with system logs:

sudo zypper history | grep python313-wheel
sudo zypper history --since='-7 days' | tail -40
sudo ausearch -m AVC,USER_AVC -ts today | tail -100
cat /proc/sys/kernel/tainted   # non-zero = tainted kernel / out-of-tree modules

Quick Triage

Run these on SLES 16 to capture the current state of python313-wheel:

rpm -q python313-wheel                              # installed NVR
rpm -V python313-wheel                              # verify shipped files
sudo zypper patch-check                    # open patches
sudo zypper lp -r SUSE-SLE-Server-16-* 2>/dev/null | head
systemctl --failed --no-pager
sudo firewall-cmd --list-all
getenforce                                  # SELinux mode
rpm -ql python313-wheel | grep -E '^/(usr/lib|etc)/systemd/(system|user)/[^/]+\.(service|socket|timer|path|target|mount)$'  # the units python313-wheel REALLY ships - NO OUTPUT means it ships no service at all, so never run systemctl on it
rpm -ql python313-wheel | grep -E '^/usr/sbin/rc[^/]+$'  # SUSE-only rcNAME compat symlink (it points at /usr/sbin/service, not at a unit): the NAME after 'rc' IS the service name - bind ships /usr/sbin/rcnamed, so the unit is named.service

Step-by-Step Diagnosis

  1. List failed systemd units.

    systemctl --failed --no-pager
  2. Tail the journal for python313-wheel and the system bus.

    sudo journalctl -xe -f --no-pager
  3. Inspect firewall posture (firewalld is the default on SLES 15+).

    sudo firewall-cmd --list-all-zones --permanent
    sudo nft list ruleset 2>/dev/null | head -50
  4. Surface SELinux denials and author a local policy module if needed.

    sudo ausearch -m AVC,USER_AVC -ts today
    sudo ausearch -m AVC -ts today | audit2allow -a -M /tmp/local-fix
    sudo semodule -i /tmp/local-fix.pp
  5. Verify python313-wheel integrity and reinstall if anything is altered.

    sudo rpm -V python313-wheel
    sudo zypper verify
    sudo zypper install --force python313-wheel
  6. Correlate findings with /var/log/zypp/history, zypper history, and SUSE advisory RHSA-2026:1902 to pin the change that introduced the regression.

Solution – Primary Fix

Apply the corrective zypper transaction referenced by SUSE advisory RHSA-2026:1902, then reload affected systemd units:

sudo zypper ref                        # refresh repos
sudo zypper -n patch                   # apply ALL open patches (recommended)
# Or target a single package:
sudo zypper -n update python313-wheel
# Unit name may differ from pkg name; check first:
rpm -q python313-wheel                           # confirm new NVR

For kernel / glibc / systemd / openssl advisories a reboot is required (or SLE Live Patching where licensed):

sudo zypper ps -s                      # services using deleted libs
sudo systemctl reboot                  # or: sudo shutdown -r now
# SUSE Live Patching (kgraft / klp) avoids reboot for kernel CVEs:
sudo zypper install -y kernel-livepatch-$(uname -r | tr - _)
klp -v patches                         # active livepatches

Need help rolling this patch across a SUSE fleet? Our IT Solutions & Services team manages SUSE patch windows with SUSE Manager / RMT and Live Patching. Get in touch for a free consultation.

Solution – Alternative Approaches

If the primary patch is not viable, choose from these:

  • Roll back via Snapper (Btrfs snapshots taken automatically before zypper transactions on SLES 16):

    sudo snapper list
    sudo snapper undochange <pre>..<post>  # diff between two snapshot numbers
    sudo snapper rollback <pre>            # boot the host into the chosen snapshot
  • Lock the package so zypper cannot upgrade it:

    sudo zypper al python313-wheel                   # add lock
    zypper ll | grep python313-wheel                 # list locks
    sudo zypper rl python313-wheel                   # remove lock
  • Install an older NVR if a regression is suspected:

    zypper se -s python313-wheel                     # show all available versions
    sudo zypper install --oldpackage python313-wheel-<older-NVR>
  • Switch SELinux to permissive briefly to confirm policy is the cause, then re-enforce:

    sudo setenforce 0
    # reproduce, capture denials, author a custom module:
    sudo ausearch -m AVC -ts recent | audit2allow -a -M mylocal
    sudo semodule -i mylocal.pp
    sudo setenforce 1
  • Where SLE Live Patching is licensed, apply kernel fixes without reboot:

    klp -v patches                         # active livepatches
    sudo zypper install -y kernel-livepatch-$(uname -r | tr - _)

Verification & Acceptance Criteria

All of these should pass after the fix:

rpm -q python313-wheel                                            # expected fixed NVR
sudo zypper patch-check                                  # 0 critical patches outstanding
sudo firewall-cmd --list-services
getenforce
sudo zypper ps -s                                        # any services still using deleted libs

The conditions described in the advisory must no longer be reported for python313-wheel across two consecutive runs.

Rollback Plan

Capture state before any change:

rpm -qa > /root/rpm-pre.txt
sudo zypper history list > /root/zypper-history-pre.txt
# Snapper takes pre/post snapshots automatically on Btrfs root.
sudo snapper create -d 'pre-patch-python313-wheel'   # explicit named snapshot
sudo snapper list | head

To revert if the patch is bad:

# Preferred on Btrfs root — boot the prior snapshot:
sudo snapper rollback <snapshot-id>
sudo systemctl reboot
# Or downgrade just the package:
sudo zypper install --oldpackage python313-wheel-<older-NVR>
# Custom security policy cleanup:
sudo semodule -r mylocal

Prevention & Hardening

Reduce the chance of this recurring on SLES 16:

  • Enable automatic patch installation:

    sudo zypper install -y zypper-automatic
    sudo systemctl enable --now zypper-automatic.timer
    # Or use YaST: yast2 online_update_configuration
  • Subscribe to sle-security-updates and watch suse.com/support/update.

  • Mirror through SUSE Manager or RMT (Repository Mirroring Tool) for controlled rollouts:

    sudo zypper install -y rmt-server rmt-cli
    sudo rmt-cli sync
    sudo rmt-cli products enable SLES/16/x86_64
  • Lock sensitive packages so they cannot be auto-upgraded:

    sudo zypper al python313-wheel
  • Ensure Snapper is enabled on the root subvolume and pre/post hooks run for every zypper transaction:

    sudo snapper -c root get-config | head
    # Default zypper plugin: /usr/lib/zypp/plugins/commit/snapper.zypp-commit-plugin
  • Monitor file integrity with AIDE:

    sudo zypper install -y aide
    sudo aide --init && sudo mv /var/lib/aide/aide.db.new /var/lib/aide/aide.db
    sudo aide --check
  • Subscribe to SUSE Live Patching so kernel CVEs can be remediated without reboot:

    sudo SUSEConnect -p sle-module-live-patching/16.0/x86_64
    sudo zypper install -y kernel-livepatch-$(uname -r | tr - _)
    klp -v patches
  • SLES 16 ships with SELinux in enforcing mode by default; review and maintain custom modules in /etc/selinux/targeted/ rather than disabling enforcement.

  • Apply CIS SUSE Linux Enterprise Server Benchmark hardening.

Issues that commonly surface alongside a python313-wheel update: zypper lock contention, systemd unit ordering cycles, SELinux AVC bursts, firewalld zone drift, and kernel taint flags. Useful triage:

sudo zypper ps -s
systemd-analyze critical-chain
sudo ausearch -m AVC -ts today | tail
sudo firewall-cmd --get-active-zones
cat /proc/sys/kernel/tainted

View all sles-16 tutorials on the Tutorials Hub →

Browse all common problems & solutions on the Tutorials Hub.

References & Further Reading

Primary reference: SUSE advisory RHSA-2026:1902 (see also SUSE bugzilla). Manual pages useful on SLES 16:

man zypper
man zypper.conf
man systemctl
man journalctl
man firewall-cmd
man snapper
man semanage
man audit2allow
man SUSEConnect
man klp

Other resources: SUSE Linux Enterprise Server 16 documentation, suse.com/security, SUSE security blog, and per-package notes in /usr/share/doc/packages/python313-wheel/ for components implicated in this advisory.