[SERVER-31425] Retrying initial shardCollection with hashed shard key after retryable error can skip moveChunk Created: 05/Oct/17 Updated: 06/Dec/22 Resolved: 17/Aug/18 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Sharding |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Jack Mulrow | Assignee: | [DO NOT USE] Backlog - Sharding Team |
| Resolution: | Duplicate | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||||||||||
| Assigned Teams: |
Sharding
|
||||||||||||||||||||||||
| Operating System: | ALL | ||||||||||||||||||||||||
| Participants: | |||||||||||||||||||||||||
| Linked BF Score: | 26 | ||||||||||||||||||||||||
| Description |
|
The last step in _configsvrShardCollection is to migrate initial chunks from empty collections to spread them across the shards, which is only done if the collection is empty and the shard key is hashed. If there is a retryable error (like a stepdown) after the step before, which is to actually shard the collection through the ShardingCatalogManager, then when the command is retried, it will early exit successfully because the collection is already sharded with the same options, which skips the moveChunk call. |
| Comments |
| Comment by Maria van Keulen [ 13/Sep/18 ] |
|
greg.mckeon Ah, I didn't see that this issue was closed as a dupe of |
| Comment by Gregory McKeon (Inactive) [ 13/Sep/18 ] |
|
maria.vankeulen we didnt do any work on this ticket - what ticket would you like to see backported? |
| Comment by Ian Whalen (Inactive) [ 28/Aug/18 ] |
|
greg.mckeon i just spotted this - maybe the 3 BFs linked here can be resolved? |