[SERVER-46408] Document behavior differences between replica set, single node replica , and single voter replica sets in Replication Architecture Guide Created: 25/Feb/20  Updated: 06/Dec/22  Resolved: 18/Jun/20

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

Type: Task Priority: Major - P3
Reporter: David Daly Assignee: Backlog - Replication Team
Resolution: Won't Do Votes: 0
Labels: perfteam-wishlist
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
is related to SERVER-48977 Add to Replication Architecture Guide... Backlog
Assigned Teams:
Replication
Participants:

 Description   

There are special cases in the code for single node replica sets and single voter replica sets. 



 Comments   
Comment by Dianna Hohensee (Inactive) [ 18/Jun/20 ]

Looks like I lost track of this. Sorry about that. I've filed a ticket for the work: SERVER-48977. Closing this one.

Comment by Dianna Hohensee (Inactive) [ 09/Mar/20 ]

judah.schvimer my exploration of this part of the system extends only to figuring out what writes with j:true write concern can survive a primary server crash, and adding a wait for no oplog holes for single voter replica set primaries to ensure behavior parity with past versions. I'm happy to add a piece about write concern variations, but I believe David Daly had more abnormalities in mind, from which I gather replication has quite a few configuration specific quirks hidden away.

Comment by Judah Schvimer [ 09/Mar/20 ]

dianna.hohensee, the replication team thinks you may be the best person to update this, given your recent work in this area. Do you think you could make this update? Do you think this belongs in the replication guide or the execution guide?

Generated at Thu Feb 08 05:11:24 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.