> Yes, the VACUUM truncation is still an issue. But statements are
> retryable, just like deadlocks.
>
> Unfo the truncation logic always kicks in or small tables of less than
> 16 blocks. It's more forgiving on bigger tables.
Oh, I didn't know that. Thanks for the info.
> Maybe we could defer the truncation on the standby in some cases.
Do we want to add this to the TODO list?
Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp