[SERVER-46425] Consider increasing wtimeout for cloneCatalogData or no timeout Created: 26/Feb/20 Updated: 12/Dec/23 |
|
| Status: | Backlog |
| Project: | Core Server |
| Component/s: | Sharding |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Bug | Priority: | Major - P3 |
| Reporter: | Randolph Tan | Assignee: | Backlog - Cluster Scalability |
| Resolution: | Unresolved | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||
| Assigned Teams: |
Cluster Scalability
|
||||||||||||||||
| Operating System: | ALL | ||||||||||||||||
| Participants: | |||||||||||||||||
| Description |
|
Current setting is majority write concern with a 60 sec wtimeout. However, the clone can potentially generate lots of writes and index builds, which can cause it to timeout waiting for replication. In the current master, _movePrimary will attempt to retry because writeConcern errors are treated as a retryable error, but since the collections were already cloned already earlier, it will get a namespace already exists error, which is not retryable and causing the entire _movePrimary command to fail. This can lead to the data ending up as orphans and eventually causing issue described in |