[SERVER-15537] Bgsync and fetch threads should behave well when node is removed from a replica set Created: 06/Oct/14  Updated: 02/Aug/18  Resolved: 07/Oct/14

Status: Closed
Project: Core Server
Component/s: Replication
Affects Version/s: None
Fix Version/s: 2.7.8

Type: Task Priority: Major - P3
Reporter: Andy Schwerin Assignee: Andy Schwerin
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Backwards Compatibility: Fully Compatible
Participants:

 Description   

These threads have very loose synchronization with the replication coordinator, and so need to be very robust to member state changes, especially to the REMOVED state.



 Comments   
Comment by Githook User [ 07/Oct/14 ]

Author:

{u'username': u'andy10gen', u'name': u'Andy Schwerin', u'email': u'schwerin@mongodb.com'}

Message: SERVER-15537 Make methods called by bgsync and fetch threads robust to node being removed from set.

If a node is removed from a replica set, or if it is not a member of its locally
stored configuration, the ReplicationCoordinator must still respond non-fatally
to method calls from the bgsync and fetch threads. It's typically OK to throw
an exception or return an error value, depending on circumstance. In this
change, we make one of each change.
Branch: master
https://github.com/mongodb/mongo/commit/863423d8cfa45a0420d8139a808cb61536a77d8b

Generated at Thu Feb 08 03:38:18 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.