[SERVER-39477] random_moveChunk workloads should tolerate failed migrations because chunk's range is in use on recipient Created: 08/Feb/19 Updated: 29/Oct/23 Resolved: 26/Mar/19 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Sharding |
| Affects Version/s: | None |
| Fix Version/s: | 4.1.10 |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Jack Mulrow | Assignee: | Jack Mulrow |
| Resolution: | Fixed | Votes: | 0 |
| Labels: | sharding-wfbf-day | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||
| Backwards Compatibility: | Fully Compatible | ||||
| Operating System: | ALL | ||||
| Sprint: | Sharding 2019-04-08 | ||||
| Participants: | |||||
| Linked BF Score: | 15 | ||||
| Description |
|
At the beginning of a chunk migration, before clearing all documents from the range to be received, the recipient shard checks no documents in that range could be in use, throwing RangeOverlapConflict and failing the migration if so. The fsm workloads random_moveChunk_broadcast_delete/update_transaction.js randomly move chunks between shards while running multi writes without the shard key (which are versioned in a transaction), so migrations can fail with this code and the tests should tolerate that. |
| Comments |
| Comment by Githook User [ 26/Mar/19 ] |
|
Author: {'email': 'jack.mulrow@mongodb.com', 'name': 'Jack Mulrow', 'username': 'jsmulrow'}Message: |