[SERVER-22875] stepdown returns ok: 1 but does not cause a stepdown if a node is draining its oplog while transitioning to primary Created: 26/Feb/16  Updated: 06/Dec/22  Resolved: 25/Apr/16

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

Type: Bug Priority: Major - P3
Reporter: Backlog - Replication Team Assignee: Backlog - Replication Team
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
Assigned Teams:
Replication
Operating System: ALL
Sprint: Repl 15 (06/03/16)
Participants:
Linked BF Score: 0

 Description   

Current bad behavior:
If a node in drain mode receives a stepdown command, the stepdown command returns ok:1 (success) but the node in drain mode does not step down. Instead, the node in drain mode finishes drain mode and then remains in state PRIMARY.
What should happen:
Upon receiving a stepdown command, a node in drain mode should interrupt drain mode, step down to SECONDARY, and return ok:1 to the stepdown command.



 Comments   
Comment by Eric Milkie [ 25/Apr/16 ]

This turned out to be a misdiagnosis of a failed jstest. The stepdown command actually did not return ok:1.

Comment by Andy Schwerin [ 29/Feb/16 ]

Please add a more detailed description.

Comment by Eric Milkie [ 27/Feb/16 ]

I would like the behavior to be that the node steps down. Alternatively, it can return an error, but that will make the proposed merged freeze/stepdown command problematic later.

Comment by Andy Schwerin [ 27/Feb/16 ]

Is that a bug? What should have happened?

Generated at Thu Feb 08 04:01:40 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.