[SERVER-36770] Provide a way to manually clean up prepared transactions Created: 20/Aug/18  Updated: 21/Feb/19  Resolved: 20/Feb/19

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

Type: Task Priority: Major - P3
Reporter: Judah Schvimer Assignee: Randolph Tan
Resolution: Won't Fix Votes: 0
Labels: prepare_backup_restore
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
related to SERVER-38060 Don't run after test hooks in resmoke... Closed
is related to TOOLS-2222 Mongodump should ignore config.transa... Closed
is related to SERVER-39699 Add a way for abortTransaction to be ... Closed
Sprint: Sharding 2019-01-14, Sharding 2019-01-28, Sharding 2019-02-11, Sharding 2019-02-25
Participants:

 Description   

If a user accidentally restores a cluster incorrectly or the server gets into a bad state, we need a way to explicitly abort a prepared transaction other than an 'abortTransaction' command from the coordinator. Preferably there would be a way to do this for all prepared transactions at once.



 Comments   
Comment by Randolph Tan [ 20/Feb/19 ]

Based on the discussion, we can just use the normal abortTransaction to abort specific transactions. For sharded transactions, this means that the DBA would need to login to the shards directly and send abortTransaction, or send abortTransaction to mongos if we implement SERVER-39699.

Comment by Judah Schvimer [ 23/Jan/19 ]

david.golden, can you please take a look at the above document?

Comment by Judah Schvimer [ 14/Jan/19 ]

renctan to investigate if we actually need this or if abortTransaction is sufficient, and if so what it should look like. Also to investigate how this works with mongos.

Generated at Thu Feb 08 04:44:02 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.