# SENTINEL: quorum not reached for failover

- **ID:** `redis/sentinel-quorum-failure`
- **Domain:** redis
- **Category:** runtime_error
- **Error Code:** `ERM`
- **Verification:** ai_generated
- **Fix Rate:** 85%

## Root Cause

Sentinel nodes could not agree on a master failure due to network partitions or insufficient sentinel instances, preventing failover.

## Version Compatibility

| Version | Status | Introduced | Deprecated |
|---------|--------|------------|------------|
| Redis 6.2.0 | active | — | — |
| Redis 7.0.0 | active | — | — |
| Redis 7.2.0 | active | — | — |

## Workarounds

1. **Ensure at least 3 sentinel instances are running with quorum set to 2. Use SENTINEL SET <master-name> quorum 2 and verify connectivity with INFO sentinel.** (90% success)
   ```
   Ensure at least 3 sentinel instances are running with quorum set to 2. Use SENTINEL SET <master-name> quorum 2 and verify connectivity with INFO sentinel.
   ```
2. **Check network connectivity between sentinels: use redis-cli -h <sentinel-ip> -p 26379 PING to each sentinel. Fix firewall rules or DNS resolution.** (85% success)
   ```
   Check network connectivity between sentinels: use redis-cli -h <sentinel-ip> -p 26379 PING to each sentinel. Fix firewall rules or DNS resolution.
   ```
3. **If a sentinel is down, restart it individually and wait for it to rejoin the quorum. Monitor with SENTINEL MASTER <master-name>.** (80% success)
   ```
   If a sentinel is down, restart it individually and wait for it to rejoin the quorum. Monitor with SENTINEL MASTER <master-name>.
   ```

## Dead Ends

- **** — Manual failover bypasses sentinel checks and can cause split-brain if the original master recovers. (75% fail)
- **** — A high quorum makes failover harder to achieve, increasing downtime during real failures. (70% fail)
- **** — Simultaneous restart can cause all sentinels to lose state and delay failover; it does not fix network issues. (80% fail)