On Thu, Dec 19, 2013 at 3:04 PM, Andres Freund <andres@2ndquadrant.com> wrote:
> On 2013-12-19 14:37:04 -0800, Sergey Konoplev wrote:
>> It was suffering from this problem on 9.2.4, mostly last couple of
>> weeks when I had to rebuild the replica almost every 3 days, and I
>> hoped it would be fixed in 9.2.6, but it is not.
>
> This actually is a separate bug from the one fixed. Not that the CONTEXT
> message isn't talking about "visible" but about "vacuum".
>
>> Yesterday I upgraded both master and replica servers to 9.2.6,
>> performed SET vacuum_freeze_table_age = 0; VACUUM; and then rebuild
>> the replica from scratch.
>
>> 2013-12-19 20:51:22 MSK 19938 @ from [vxid:1/0 txid:0] [] WARNING:
>> page 14833 of relation base/16436/3321003988 is uninitialized
>> 2013-12-19 20:51:22 MSK 19938 @ from [vxid:1/0 txid:0] [] CONTEXT:
>> xlog redo vacuum: rel 1663/16436/3321003988; blk 38538,
>> lastBlockVacuumed 0
>
> Heikki, that's the redo routine for XLOG_BTREE_VACUUM, you probably know
> that code best atm, and it might influence on how the new split and
> deletion code works...
I just wonder if it might be wal_level = hot_standby problem only.
What if I switch it to just archive? Wont this problem gone?
--
Kind regards,
Sergey Konoplev
PostgreSQL Consultant and DBA
http://www.linkedin.com/in/grayhemp
+1 (415) 867-9984, +7 (901) 903-0499, +7 (988) 888-1979
gray.ru@gmail.com
