-
Type:
Improvement
-
Resolution: Unresolved
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
Catalog and Routing
-
2
-
🟥 DDL
-
None
-
None
-
None
-
None
-
None
-
None
The execution and error handling of the collmod coordinator could be simplified if we stored the collection and sharding information on the coordinator document when we calculate it the first time. Additionally, providing separate phases for the release of the critical sections and re-enablement of chunk migrations would simplify the execution (we could remove the release of the critical section from the collmod participant, for example). And finally, adding a reason to the collmod critical section will ensure we don't have reason conflicts in the future and simplify debugging.
- is related to
-
SERVER-125921 Sharded timeseries collMod can abort without releasing participant critical sections
-
- Closed
-
-
SERVER-110653 Investigate removing/reducing the CRUD block during granularity changes on timeseries
-
- Closed
-
- related to
-
SERVER-133502 Sharded time-series collMod can return success while leaving metadata inconsistent if the DB-primary participant fails
-
- Backlog
-