[SERVER-72178] Index cache does not consider finer grained refreshes Created: 16/Dec/22  Updated: 27/Oct/23  Resolved: 23/Mar/23

Status: Closed
Project: Core Server
Component/s: Sharding
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: Marcos José Grillo Ramirez Assignee: Sergi Mateo Bellido
Resolution: Works as Designed Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
is depended on by SERVER-49701 Make enableFinerGrainedCatalogCacheRe... Blocked
Assigned Teams:
Sharding EMEA
Operating System: ALL
Sprint: Sharding EMEA 2023-03-20, Sharding EMEA 2023-04-03
Participants:

 Description   

enableFinerGrainedCatalogCacheRefresh is a server parameter used to prevent queries from waiting for a catalog refresh if the information is stale. Currently this is not being considered in the index cache at all.

We should check which should be the index cache behavior if the parameter is enabled.



 Comments   
Comment by Sergi Mateo Bellido [ 23/Mar/23 ]

I believe that fine-grained refreshes don't make sense for the index cache since the index information is global (in other words, it is not per shard). Thus, if a router knows that it is stale we should always refresh, independently of which shards are going to be targeted.

Comment by Marcos José Grillo Ramirez [ 13/Feb/23 ]

Before re-enabling the finer grain, this should be taken into consideration.

Generated at Thu Feb 08 06:21:03 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.