Re: Bug: Unreferenced temp tables disables vacuum to update xid

Tom Lane wrote:

> regression=# cluster foo_pkey on foo;
> CLUSTER
> regression=# select relname, relkind, relfrozenxid from pg_class order by oid desc limit 6;
>         relname        | relkind | relfrozenxid 
> -----------------------+---------+--------------
>  pg_toast_707231_index | i       |            0
>  pg_toast_707231       | t       |       119424
>  foo_pkey              | i       |            0
>  foo                   | r       |   4195086720
>  foo_f1_seq            | S       |            0
>  xmlview5              | v       |            0
> (6 rows)
> 
> So something is out of whack in CLUSTER.  However it only seems to be
> broken in HEAD, so I'm not sure this helps to explain the original
> report.  (Speculation: this is related to the rewrite to make CLUSTER
> MVCC-safe?)

Right ... see copy_heap_data --- it sets FreezeXid as relfrozenxid.

If we were to scan each tuple as it is inserted, we could store a higher
relfrozenxid, but I doubt we want to do that.

Perhaps what we could do is take the relfrozenxid from the old relation
and copy it over, if it's later than FreezeXid?

-- 
Alvaro Herrera                                http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.