Re: deadlock of lock-waits (on transaction and on tuple) using same update statement - Mailing list pgsql-general

From Andrej Vanek
Subject Re: deadlock of lock-waits (on transaction and on tuple) using same update statement
Date
Msg-id CAFNFRyF-URJAwbJCdHM6oOBvuHOxYkRWKBpfikF8Y7M9QBwWPQ@mail.gmail.com
Whole thread Raw
In response to Re: deadlock of lock-waits (on transaction and on tuple) using same update statement  (Bill Moran <wmoran@potentialtech.com>)
Responses Re: deadlock of lock-waits (on transaction and on tuple) using same update statement  (Andrej Vanek <andrej.vanek.sk@gmail.com>)
List pgsql-general
Hi Bill,

thanks for your answer.

most often caused by something earlier in the transactions
 need all of the statements in each transaction

It would be great if we could reveal an application error.

Whole transactions I've already posted (in postgres log: log_min_duration_statement=0).
Nothing suspicious yet: 
- both sessions COMMIT/ROLLBACK before BEGIN
- both sessions run the same SQL statements
- deadlock: FIRST statement of both transactions
- deadlock: update single table, same row, column without any constraints (WHY?)
- no statements of other sessions executed in between

Any idea?

thanks, Andrej

pgsql-general by date:

Previous
From: Xiang Gan
Date:
Subject: Re: Installing Postgresql on Linux Friendlyarm
Next
From: Andrej Vanek
Date:
Subject: Re: deadlock of lock-waits (on transaction and on tuple) using same update statement