Horton

Re: test_decoding assertion failure for the loss of top-sub transaction relationship

Kyotaro Horiguchi <horikyota.ntt@gmail.com>

View thread View on postgr.es
From: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
To: kuroda.hayato@fujitsu.com
Cc: pgsql-hackers@lists.postgresql.org, osumi.takamichi@fujitsu.com
Date: 2022-09-02T04:16:05Z
Lists: pgsql-hackers 
Good catch, and thanks for the patch!

At Fri, 2 Sep 2022 01:08:04 +0000, "kuroda.hayato@fujitsu.com" <kuroda.hayato@fujitsu.com> wrote in 
> PSA patch that fixes the failure.
> This adds pairs of sub-top transactions to the SnapBuild, and it will be serialized and restored.
> The pair will be checked when we mark the ReorderBufferTXN as RBTXN_HAS_CATALOG_CHANGES.

A commit record has all subtransaction ids and SnapBuildCommitTxn()
already checks if every one has catalog changes before checking the
top transaction's catalog changes. So, no need to record top-sub
transaction relationship to serialized snapshots. If any of the
subtransactions has catalog changes, the commit contains catalog
changes.

regards.

-- 
Kyotaro Horiguchi
NTT Open Source Software Center

Commits

  1. Fix executing invalidation messages generated by subtransactions during decoding.

  2. Fix assertion failures while processing NEW_CID record in logical decoding.