[SERVER-57230] mongos doesn't distribute secondary reads evenly on v4.4 or above Created: 26/May/21  Updated: 31/May/22  Resolved: 31/May/22

Status: Closed
Project: Core Server
Component/s: None
Affects Version/s: 4.4.0
Fix Version/s: None

Type: Bug Priority: Major - P3
Reporter: Tyler Seip (Inactive) Assignee: Vojislav Stojkovic
Resolution: Duplicate Votes: 2
Labels: servicearch-q4-2021, servicearch-wfbf-day
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
depends on SERVER-65131 Disable opportunistic read targeting ... Closed
Duplicate
duplicates SERVER-57232 Increase in slow queries in Connectio... Closed
is duplicated by SERVER-65131 Disable opportunistic read targeting ... Closed
Related
Operating System: ALL
Sprint: Service Arch 2021-06-28, Service Arch 2021-07-12, Service Arch 2021-08-09, Service Arch 2021-08-23, Service Arch 2021-10-04, Service Arch 2022-1-24, Service Arch 2022-2-07, Service Arch 2022-2-21, Service Arch 2022-03-07, Service Arch 2022-03-21, Service Arch 2022-04-04
Participants:
Case:
Story Points: 5

 Description   

In v4.2, mongos correctly balanced outgoing read requests to secondary nodes by randomizing the connection order when we issued requests here. In v4.4, the introduction of either hedged reads, opportunistic reads, or changes to ConnectionPool have made it such that one of these connections now consistently resolves first, effectively starving one of the secondaries of any reads and unbalancing our read requests. In this ticket, create a unit or integration test to capture this behavior and fix it.



 Comments   
Comment by Vojislav Stojkovic [ 31/May/22 ]

Fixed by SERVER-65131.

Comment by Lauren Lewis (Inactive) [ 28/Oct/21 ]

De-prioritized, not currently working. Will revisit & keeping assigned to tyler.seip

Generated at Thu Feb 08 05:41:19 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.