fix(core): Prevent Redis connection recovery from being missed

[ivov]

Summary

RedisClientService debounces all event emissions by 1s, and the reconnection retry interval is also 1s. When a connection drops, retryStrategy queues a debounced connection-lost event. Meanwhile, ioRedis waits 1s and reconnects. So the debounce timer and the ready event race each other. Typically the debounce wins and sets lostConnection = true before ready checks it, but under load ready can fire first, so it finds lostConnection still as false and skips recovery, then the late connection-lost flips the flag to true, where it stays forever.

This PR sets lostConnection = true synchronously in retryStrategy to bypass the debounce. The debounced listener now only logs.

Related Linear tickets, Github issues, and Community forum posts

https://linear.app/n8n/issue/CAT-2721
Closes #27753

Review / Merge checklist

• PR title and summary are descriptive. (conventions)
• Docs updated or follow-up ticket created.
• Tests included.
• PR Labeled with Backport to Beta, Backport to Stable, or Backport to v1 (if the PR is an urgent fix that needs to be backported)

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[cubic-dev-ai]

No issues found across 2 files

Architecture diagram

sequenceDiagram
    participant Redis as ioRedis Library
    participant RCS as RedisClientService
    participant Emitter as Internal Emitter (Debounced)

    Note over Redis, RCS: Connection Failure
    Redis->>RCS: retryStrategy() callback invoked

    rect rgb(240, 240, 240)
        Note right of RCS: NEW: Atomic State Update
        RCS->>RCS: Set lostConnection = true
    end

    RCS->>Emitter: emit('connection-lost', timeout)
    
    Note over Emitter: Debounce Timer Starts (1s)

    Note over Redis, RCS: Connection Re-established
    Redis->>RCS: Event: 'ready'

    alt is lostConnection == true
        Note right of RCS: Recovery Path Triggered
        RCS->>RCS: internalRecover()
        RCS->>Emitter: emit('connection-recovered')
        RCS->>RCS: Set lostConnection = false
    else CHANGED: Previous Race Condition (if ready fired before debounce)
        Note right of RCS: Recovery skipped (Bug fixed)
    end

    Note over Emitter: Debounce Timer Expires
    Emitter-->>RCS: Listener: 'connection-lost' fires
    
    rect rgb(240, 240, 240)
        Note right of RCS: CHANGED: Only logs warning now
        RCS->>RCS: logger.warn(...)
    end

Loading

[github-actions]

Performance Comparison

Comparing current → latest master → 14-day baseline

docker-stats

Metric	Current	Latest Master	Baseline (avg)	vs Master	vs Baseline	Status
docker-image-size-n8n	1269.76 MB	1269.76 MB	1269.76 MB (σ 0.00)	+0.0%	+0.0%	—
docker-image-size-runners	386.00 MB	386.00 MB	388.00 MB (σ 3.46)	+0.0%	-0.5%	✅

Memory consumption baseline with starter plan resources

Metric	Current	Latest Master	Baseline (avg)	vs Master	vs Baseline	Status
memory-rss-baseline	287.07 MB	287.07 MB	281.78 MB (σ 34.50)	+0.0%	+1.9%	✅
memory-heap-used-baseline	114.53 MB	114.53 MB	113.09 MB (σ 1.15)	+0.0%	+1.3%	⚠️

Idle baseline with Instance AI module loaded

Metric	Current	Latest Master	Baseline (avg)	vs Master	vs Baseline	Status
instance-ai-heap-used-baseline	186.41 MB	186.41 MB	186.41 MB (< 3 samples)	+0.0%	+0.0%	—
instance-ai-rss-baseline	343.76 MB	343.76 MB	343.76 MB (< 3 samples)	+0.0%	-0.0%	—

How to read this table

• Current: This PR's value (or latest master if PR perf tests haven't run)
• Latest Master: Most recent nightly master measurement
• Baseline: Rolling 14-day average from master
• vs Master: PR impact (current vs latest master)
• vs Baseline: Drift from baseline (current vs rolling avg)
• Status: ✅ within 1σ | ⚠️ 1-2σ | 🔴 >2σ regression