[SERVER-59875] Ability to Use Replica Set Tags with initialSyncSourceReadPreference Created: 10/Sep/21  Updated: 05/Dec/22

Status: Open
Project: Core Server
Component/s: None
Affects Version/s: 4.2.7, 4.4.8, 5.0.2
Fix Version/s: None

Type: Improvement Priority: Major - P3
Reporter: Diego Rodriguez (Inactive) Assignee: Backlog - Replication Team
Resolution: Unresolved Votes: 1
Labels: pmr
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
Assigned Teams:
Replication
Participants:

 Description   

Hi Team,

In SERVER-38731 we introduced the ability to specify a parameter called initialSyncSourceReadPreference that allows one of the following options:

  • primary
  • primaryPreferred (Default for voting replica set members)
  • secondary
  • secondaryPreferred
  • nearest (Default for newly added or non-voting replica set members)

I open this ticket to propose an improvement to this, allowing tag sets to be used for this parameter with the ability to configure replica set tags in order of preference just like when working with connection strings.

The reasoning is that this would allow users to control which replica set members are used as initial sync sources and avoid overloading members of the replica set that participate in low latency and/or critical operations. As an example, a user might want the initial sync overhead to be handled by Analytics nodes (MongoDB Atlas is a perfect example) or hidden nodes that do not interfere with low latency and/or critical workload.

This would affect the following versions:

  • v4.2.7+
  • v.4.4+
  • v5.0+

Regards
Diego


Generated at Thu Feb 08 05:48:23 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.