[AutoPR- Security] Patch openssh for CVE-2026-73283, CVE-2026-73282, CVE-2026-73281 [MEDIUM]

[Azure Linux Security Servicing Account (azurelinux-security)]

Auto Patch openssh for CVE-2026-73283, CVE-2026-73282, CVE-2026-73281.

Autosec pipeline run -> https://dev.azure.com/mariner-org/mariner/_build/results?buildId=1183633&view=results

CVE-2026-73282 : Single Patch Backporter Pipeline Run -> https://dev.azure.com/mariner-org/mariner/_build/results?buildId=1183638&view=results
CVE-2026-73281 : Single Patch Backporter Pipeline Run -> https://dev.azure.com/mariner-org/mariner/_build/results?buildId=1183639&view=results

Merge Checklist

All boxes should be checked before merging the PR (just tick any boxes which don't apply to this PR)

• The toolchain has been rebuilt successfully (or no changes were made to it)
• The toolchain/worker package manifests are up-to-date
• Any updated packages successfully build (or no packages were changed)
• Packages depending on static components modified in this PR (Golang, *-static subpackages, etc.) have had their Release tag incremented.
• Package tests (%check section) have been verified with RUN_CHECK=y for existing SPEC files, or added to new SPEC files
• All package sources are available
• cgmanifest files are up-to-date and sorted (./cgmanifest.json, ./toolkit/scripts/toolchain/cgmanifest.json, .github/workflows/cgmanifest.json)
• LICENSE-MAP files are up-to-date (./LICENSES-AND-NOTICES/SPECS/data/licenses.json, ./LICENSES-AND-NOTICES/SPECS/LICENSES-MAP.md, ./LICENSES-AND-NOTICES/SPECS/LICENSE-EXCEPTIONS.PHOTON)
• All source files have up-to-date hashes in the *.signatures.json files
• sudo make go-tidy-all and sudo make go-test-coverage pass
• Documentation has been updated to match any changes to the build system
• Ready to merge

---

Summary

What does the PR accomplish, why was it needed?

• Auto Patch openssh for CVE-2026-73283 (LOW), CVE-2026-73282 (MEDIUM), CVE-2026-73281 (LOW).

Change Log

• CVE-2026-73283
• CVE-2026-73282
• CVE-2026-73281

Does this affect the toolchain?

YES/NO

Associated issues

• N/A

Links to CVEs

• https://nvd.nist.gov/vuln/detail/CVE-2026-73283
• https://nvd.nist.gov/vuln/detail/CVE-2026-73282
• https://nvd.nist.gov/vuln/detail/CVE-2026-73281

Test Methodology

• Pipeline build id: Buddy Build URL

[azure-pipelines]

Azure Pipelines:
There may be pipelines that require an authorized user to comment /azp run to run.

[v-sushilsati]

The AI-generated openssh.spec changes were incorrect and caused the build to fail. The spec was corrected as follows:
• Incremented the OpenSSH package release from 9 to 10.
• Renumbered the new CVE patches to follow the existing CVE patch sequence:
o Patch966 → Patch416 for CVE-2026-73281
o Patch967 → Patch417 for CVE-2026-73282
o Patch968 → Patch418 for CVE-2026-73283

• Moved the three patch applications from the beginning of %prep to after Patch415.
• Added a 9.8p1-10 changelog entry covering CVE-2026-73281, CVE-2026-73282, and CVE-2026-73283.

CVE-2026-73281
Backport: Yes (AI patch is not ok)
OpenSSH 9.8p1 is within the affected range (<10.5)
The vulnerable file (ssh-agent.c) and its associated logic are present in source.
The original AI-generated patch was not directly suitable for OpenSSH 9.8p1. The following code was removed
+ } else if (strcmp(name, "query") == 0) {
+ replied = success = process_ext_query(e);
The "query" SSH-agent extension and process_ext_query() function were introduced in a later OpenSSH version and do not exist in OpenSSH 9.8p1. This functionality is unrelated to CVE-2026-73281 and should not be included in the 9.8p1 backport.
the remaining logic matches the relevant upstream CVE-2026-73281 fix.
The Upstream-reference in the patch header is correct.

CVE-2026-73282

Backported: yes (AI patch is ok).
The affected code is present in OpenSSH 9.8p1.
The AI-generated backport matches the relevant upstream fix.
The Upstream-reference in the patch header is correct.
Changes made in ssh.c:
• Added struct rfwd_confirm_ctx to store a stable remote-forward index.
• Replaced the callback context pointer into options.remote_forwards with the new independently allocated context.
• Securely freed the callback context with freezero().

CVE-2026-73283
Backported: yes (AI patch is ok)
The affected code is present in OpenSSH 9.8p1.
The AI-generated backport matches the relevant upstream fix.
The Upstream-reference in the patch header is correct.

Changes made in serverloop.c:
• Added auth_opts->restricted to the tunnel-forwarding authorization check.
• Tunnel forwarding is now rejected when an authorized_keys entry uses the restrict option.

Note:
Due to a circular dependency involving a toolchain package, the buddy build PR Pipelines - Run PR-18465+openssh+unknown could not be completed successfully.

The patches have been successfully applied, and local build has been passed.