Re: [Proposal] Global temporary tables - Mailing list pgsql-hackers

From 曾文旌
Subject Re: [Proposal] Global temporary tables
Date
Msg-id A77FCEA7-FCD9-43D9-969D-7E1EAD83F9BF@alibaba-inc.com
Whole thread Raw
In response to Re: [Proposal] Global temporary tables  (Prabhat Sahu <prabhat.sahu@enterprisedb.com>)
Responses Re: [Proposal] Global temporary tables  (Prabhat Sahu <prabhat.sahu@enterprisedb.com>)
List pgsql-hackers


2020年4月7日 下午6:22,Prabhat Sahu <prabhat.sahu@enterprisedb.com> 写道:

Thanks for review.
This parameter should support all types of writing of the bool type like parameter autovacuum_enabled.
So I fixed in global_temporary_table_v24-pg13.patch.

Thank you Wenjing for the new patch with the fix and the "VACUUM FULL GTT" support.
I have verified the above issue now its resolved.

Please check the below findings on VACUUM FULL.

postgres=# create global temporary table  gtt(c1 int) on commit preserve rows;
CREATE TABLE
postgres=# vacuum FULL ;
WARNING:  global temp table oldest FrozenXid is far in the past
HINT:  please truncate them or kill those sessions that use them.
VACUUM


This is expected,
This represents that the GTT FrozenXid is the oldest in the entire db, and dba should vacuum the GTT if he want to push the db datfrozenxid.
Also he can use function pg_list_gtt_relfrozenxids() to check which session has "too old” data and truncate them or kill the sessions.




--

With Regards,
Prabhat Kumar Sahu
EnterpriseDB: http://www.enterprisedb.com


Attachment

pgsql-hackers by date:

Previous
From: Kyotaro Horiguchi
Date:
Subject: Re: [HACKERS] Restricting maximum keep segments by repslots
Next
From: Masahiko Sawada
Date:
Subject: Re: pg_stat_statements issue with parallel maintenance (Was Re: WALusage calculation patch)