[FIXED] Avoid nil deref on mirror state restore

[MauriceVanVeen]

Guard mset.mirror against nil when restoring sequence/lag state after a failed persist in processInboundMirrorMsg, preventing a potential nil deref if the mirror was removed concurrently.

Signed-off-by: Maurice van Veen github@mauricevanveen.com

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: d8eb3e67b5

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 1fb9f1c222

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[MauriceVanVeen]

Need to add a test..

[MauriceVanVeen]

Have repurposed this PR. It now only contains the mirror nil deref.

[MauriceVanVeen]

@claude, brief review.

[synadia-claude-reviewer]

Claude finished @MauriceVanVeen's task in 49s —— View job

---

Brief Review

Reviewing: server/stream.go nil guard in processInboundMirrorMsg

---

What the fix does

processInboundMirrorMsg drops mset.mu at line 3248 to perform the persist (propose or processJetStreamMsg). If an errLastSeqMismatch is returned it re-acquires the lock and writes back the saved sequence/lag values. Between the two lock windows a concurrent goroutine can set mset.mirror = nil (e.g., mirror removal), causing a nil deref. The two added nil guards close that window.

Looks correct. The pattern is consistent with how other post-unlock mirror accesses are already guarded in this file (e.g., setMirrorErr at line 3300 has the same mset.mirror != nil pattern).

A couple of observations:

1. return false when nil in the first branch (line 3282–3283): If the mirror was removed concurrently there is nothing to retry, so silently returning false is the right call.

2. retryMirrorConsumer() still called when nil in the second branch (line 3291): retryMirrorConsumer → cancelMirrorConsumer already has its own nil guard, and setupMirrorConsumer will re-create mset.mirror if needed, so this is also safe.

3. Missing test — already flagged by the author. Worth adding a concurrent-removal test to prevent regression.

Overall: minimal, safe, and matches existing defensive patterns in the file. LGTM pending a test.

---