Send close_notify after user_canceled even under quiet shutdown

[night1rider]

Clear quiet shutdown across the shutdown call in wolfSSL_SendUserCanceled so the RFC 9846 paired close_notify is emitted, and cover it in test_wolfSSL_SendUserCanceled.

[wolfSSL-Fenrir-bot]

Fenrir Automated Review — PR #11264

Scan targets checked: wolfssl-bugs, wolfssl-src

Findings: 2
2 finding(s) posted as inline comments (see file-level comments below)

This review was generated automatically by Fenrir. Reported findings require changes before merge.

[wolfSSL-Bot]

Can one of the admins verify this patch?

[wolfSSL-Fenrir-bot]

Fenrir Automated Review — PR #11264

Scan targets checked: wolfssl-bugs, wolfssl-src

Findings: 1
1 finding(s) posted as inline comments (see file-level comments below)

This review was generated automatically by Fenrir. Reported findings require changes before merge.

[github-actions]

MemBrowse Memory Report

gcc-arm-cortex-m3

• FLASH: .text +92 B (+0.1%, 125,867 B / 262,144 B, total: 48% used)

gcc-arm-cortex-m4

• FLASH: .text +64 B (+0.0%, 205,021 B / 262,144 B, total: 78% used)

gcc-arm-cortex-m4-dtls13

• FLASH: .text +128 B (+0.1%, 186,748 B / 1,048,576 B, total: 18% used)

gcc-arm-cortex-m4-openssl-compat

• FLASH: .text +64 B (+0.0%, 783,268 B / 1,048,576 B, total: 75% used)

gcc-arm-cortex-m4-pq

• FLASH: .text +64 B (+0.0%, 301,704 B / 1,048,576 B, total: 29% used)

gcc-arm-cortex-m4-rsa-only

• FLASH: .text +64 B (+0.0%, 332,512 B / 1,048,576 B, total: 32% used)

gcc-arm-cortex-m4-tls12

• FLASH: .text +64 B (+0.1%, 126,643 B / 262,144 B, total: 48% used)

gcc-arm-cortex-m4-tls13

• FLASH: .text +128 B (+0.1%, 242,255 B / 262,144 B, total: 92% used)

gcc-arm-cortex-m7

• FLASH: .text +64 B (+0.0%, 205,021 B / 262,144 B, total: 78% used)

gcc-arm-cortex-m7-pq

• FLASH: .text +64 B (+0.0%, 302,664 B / 1,048,576 B, total: 29% used)

gcc-arm-cortex-m7-tls13

• FLASH: .text +64 B (+0.0%, 242,255 B / 262,144 B, total: 92% used)
  No memory changes detected for:
• gcc-arm-cortex-m0plus
• gcc-arm-cortex-m4-baremetal
• gcc-arm-cortex-m4-crypto-only
• gcc-arm-cortex-m4-min-ecc
• gcc-arm-cortex-m4-pkcs7
• gcc-arm-cortex-m4-sp-math
• linuxkm-pie
• linuxkm-standard
• stm32-sim-stm32h753

[Copilot]

Pull request overview

This PR ensures wolfSSL_SendUserCanceled() still emits the RFC 9846-required paired close_notify even when quiet shutdown is enabled, and expands the API test to cover the quiet-shutdown scenario.

Changes:

• Temporarily clears ssl->options.quietShutdown around the internal wolfSSL_shutdown() call in wolfSSL_SendUserCanceled(), then restores it.
• Extends test_wolfSSL_SendUserCanceled() to run each protocol variant twice (normal + quiet shutdown where available) and validate the expected alert history behavior.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
src/ssl_api_rw.c	Clears quiet shutdown across wolfSSL_shutdown() inside wolfSSL_SendUserCanceled() to guarantee the paired close_notify is sent.
tests/api.c	Adds a quiet-shutdown test pass for wolfSSL_SendUserCanceled() and asserts the close-notify behavior.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

🔵 Needs a closer look

It changes core shutdown behavior in a security-critical API path (wolfSSL_shutdown / alert flushing), so it warrants final human review despite targeted test coverage.

Review details

• Files reviewed: 2/2 changed files
• Comments generated: 0 new
• Review effort level: Lite

[wolfSSL-Fenrir-bot]

Fenrir Automated Review — PR #11264

Scan targets checked: wolfssl-bugs, wolfssl-src

Findings: 3
3 finding(s) posted as inline comments (see file-level comments below)

This review was generated automatically by Fenrir. Reported findings require changes before merge.

[wolfSSL-Fenrir-bot]

Quiet shutdown flushes unrelated buffered data · API contract violations

wolfSSL_set_shutdown() can set sentNotify without sending an alert. With application data buffered after WANT_WRITE, this branch transmits that data during quiet shutdown. Known #10650 instead covers a missing required close_notify.

Related known finding #10650 (similar but distinct): Both affect quiet-shutdown handling in wolfSSL_shutdown, but the candidate's faulting operation is flushing pre-existing buffered application data after sentNotify is set, whereas #10650 omits a required close_notify. Their root causes and required patches differ: state-gating the flush versus sending the closure alert.

Suggested fix: Gate the quiet-path flush on dedicated state set only when wolfSSL_SendUserCanceled() buffers its paired close_notify.
Basis: OpenSSL SSL_CTX_set_quiet_shutdown(3), Quiet Shutdown requires SSL_shutdown() to succeed immediately without transmitting an alert.

[wolfSSL-Fenrir-bot]

Permanent flush failures remain untested · Missing edge-case coverage on a function the PR also changed

The test restores a successful callback before retrying, leaving the new branch that preserves fatal SendBuffered() errors untested. A regression converting permanent write failure to success would pass.

Suggested fix: Add a quiet-shutdown retry whose callback returns WOLFSSL_CBIO_ERR_GENERAL, then assert the fatal return and stored error.
Basis: The wolfSSL_shutdown() API contract requires errors to return WOLFSSL_FATAL_ERROR with the reason available through wolfSSL_get_error().

[wolfSSL-Fenrir-bot]

Test never verifies the user_canceled alert · Weak or missing assertions

test_wolfSSL_SendUserCanceled() checks only the record count and final close_notify, so two close_notify records satisfy it without a user_canceled. Unlike known #10650's runtime defect, this is a distinct test-oracle gap.

Suggested fix: Capture both alerts and assert user_canceled then close_notify, at minimum for every TLS/DTLS 1.3 quiet/WANT_WRITE variant.
Basis: RFC 9846 Section 6.1 requires a user_canceled alert to be followed by a close_notify.