# MongoServerError: balancer round migration failed: chunk migration failed for namespace mydb.mycoll

- **ID:** `mongodb/balancer-round-migration-failure`
- **Domain:** mongodb
- **Category:** runtime_error
- **Verification:** ai_generated
- **Fix Rate:** 85%

## Root Cause

The balancer failed to migrate a chunk due to a temporary network issue or a conflicting operation on the source or target shard.

## Version Compatibility

| Version | Status | Introduced | Deprecated |
|---------|--------|------------|------------|
| MongoDB 7.0 | active | — | — |
| MongoDB 6.0 | active | — | — |
| MongoDB 8.0 (early access) | active | — | — |

## Workarounds

1. **Check balancer state and retry migration after resolving any conflicting operations (e.g., long-running queries or index builds). Use sh.status() to monitor balancer activity.** (75% success)
   ```
   Check balancer state and retry migration after resolving any conflicting operations (e.g., long-running queries or index builds). Use sh.status() to monitor balancer activity.
   ```
2. **Increase balancer window or reduce chunk size to make migrations more manageable: use sh.setBalancerState(true) and adjust chunk size via sh._setBalancerChunkSize(64) (in MB).** (80% success)
   ```
   Increase balancer window or reduce chunk size to make migrations more manageable: use sh.setBalancerState(true) and adjust chunk size via sh._setBalancerChunkSize(64) (in MB).
   ```
3. **Force chunk migration manually using moveChunk with a specific shard key range and wait for completion: sh.moveChunk('mydb.mycoll', { shardKey: value }, 'targetShard').** (85% success)
   ```
   Force chunk migration manually using moveChunk with a specific shard key range and wait for completion: sh.moveChunk('mydb.mycoll', { shardKey: value }, 'targetShard').
   ```

## Dead Ends

- **** — moveChunk may fail with same error if underlying issue persists, and can disrupt ongoing migrations. (60% fail)
- **** — Stopping balancer is a temporary pause, not a fix; the root cause (e.g., network blip) remains. (70% fail)
- **** — Restart often doesn't resolve underlying connectivity or operation conflicts. (50% fail)