jianliu commented on pull request #11587: URL: https://github.com/apache/shardingsphere/pull/11587#issuecomment-898447941
@tristaZero @TeslaCN Sorry for the deviation of the previous reason analysis. I have add a lot of logs to reproduce error. and I have added a comment to the issue like below. I have made a mistake when analyz the reason because of missing an important info,that between pre tps and next tps. ``` #first statements set autocommit=false -- sql1:thread1 select * from tablexx where sharding_key=? -- sql2:thread2 update tablexx set x=x where sharding_key=? -- sql3:thread3 commit -- sql4:thread4 context.flush close resources because there is no transaction yet after commit #the second sql by framework (hikari) set autocommit=true-- sql1:thread10 this sql is request by datasource framework when connection is close(back to pool) context.flush close resources because there is no transaction at all #the thrid statement set autocommit=false -- sql1:thread6 select * from tablexx where sharding_key=? -- sql2:thread7 ``` Here is the logs:  The second sql (setAutoCommit(true) between the first and the thrid statement send by framework cause this erorr,because it removes the third sql(setAutoCommit:false) `s connectionPostProcessor,and lead this error finally. The PR could resolve this error too. or you could make close resources and add resources synchronzied,it`s alose resolve this issue too. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
