[SERVER-33430] Do not update in-memory Session state or on disk txn table for individual writes in ongoing autocommit:false transactions Created: 21/Feb/18  Updated: 02/Mar/18  Resolved: 02/Mar/18

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

Type: Task Priority: Major - P3
Reporter: Spencer Brody (Inactive) Assignee: Siyuan Zhou
Resolution: Duplicate Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Duplicate
duplicates SERVER-32320 Push applyOps oplog entry generation ... Closed
Related
related to SERVER-33431 Fail retries of writes in multi-state... Closed
is related to SERVER-33477 Update session entry with _id upsert Closed
is related to SERVER-33354 Add snapshot read support for additio... Closed
Sprint: Repl 2018-02-26, Repl 2018-03-12
Participants:

 Description   

We will only support retryablility for abort and commit commands, but not for individual writes in a multi-statement transaction, so we need to bypass the in-memory session state change and the on-disk transaction table change for individual writes in a multi-statement transaction. Abort and commit commands need those changes in SERVER-33501.



 Comments   
Comment by Siyuan Zhou [ 02/Mar/18 ]

The ticket has been done as part of SERVER-32320. Closing as a dup.

Comment by Gregory McKeon (Inactive) [ 02/Mar/18 ]

siyuan.zhou Can we close this as a dupe?

Comment by Siyuan Zhou [ 26/Feb/18 ]

I believe this is part of Matthew's work SERVER-32320 to push applyOps oplog entry generation into logOp on transaction commit. In opObserverImpl, we return earlier if it's a multi-statement transaction.

Comment by Andy Schwerin [ 25/Feb/18 ]

Could you add a little more description, please?

Generated at Thu Feb 08 04:33:23 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.