Re: Possible bug in logical replication.

Поиск
Список
Период
Сортировка
От Masahiko Sawada
Тема Re: Possible bug in logical replication.
Дата
Msg-id CAD21AoDEuiuPPRH+bH-9=j1mNEhY-2AXKVpzgNmS_NX5tg4QBg@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Possible bug in logical replication.  (Alvaro Herrera <alvherre@2ndquadrant.com>)
Список pgsql-hackers
On Fri, Jun 15, 2018 at 5:06 AM, Alvaro Herrera
<alvherre@2ndquadrant.com> wrote:
> Hello
>
> Can somebody (Arseny, Konstantin, horiguti, Sawada) please confirm that
> Michaël's commit fixes the reported bug?
>

I don't confirm that commit deeply yet but I have confirmed that the
reported bug is fixed using the following test case which led an
assertion failure. And this test case is the same as the previous
report[1].

postgres(1:128004)=# select pg_create_logical_replication_slot('s1',
'pgoutput');
2018-06-15 18:04:31.301 JST [128004] LOG:  logical decoding found
consistent point at 0/1645388
2018-06-15 18:04:31.301 JST [128004] DETAIL:  There are no running transactions.
2018-06-15 18:04:31.301 JST [128004] STATEMENT:  select
pg_create_logical_replication_slot('s1', 'pgoutput');
 pg_create_logical_replication_slot
------------------------------------
 (s1,0/16453C0)
(1 row)

postgres(1:128004)=# select pg_switch_wal();
 pg_switch_wal
---------------
 0/16453D8
(1 row)

postgres(1:128004)=# select pg_switch_wal();
 pg_switch_wal
---------------
 0/2000000
(1 row)

postgres(1:128004)=# select pg_replication_slot_advance('s1', '0/2000000');
2018-06-15 18:04:31.338 JST [128004] LOG:  starting logical decoding
for slot "s1"
2018-06-15 18:04:31.338 JST [128004] DETAIL:  Streaming transactions
committing after 0/16453C0, reading WAL from 0/1645388.
2018-06-15 18:04:31.338 JST [128004] STATEMENT:  select
pg_replication_slot_advance('s1', '0/2000000');
2018-06-15 18:04:31.339 JST [128004] LOG:  logical decoding found
consistent point at 0/1645388
2018-06-15 18:04:31.339 JST [128004] DETAIL:  There are no running transactions.
2018-06-15 18:04:31.339 JST [128004] STATEMENT:  select
pg_replication_slot_advance('s1', '0/2000000');
 pg_replication_slot_advance
-----------------------------
 (s1,0/2000000)
(1 row)

postgres(1:128004)=# create table a (c int);
select pg_replication_slot_advance('s1', pg_current_wal_lsn());
CREATE TABLE
postgres(1:128004)=# select pg_replication_slot_advance('s1',
pg_current_wal_lsn());
2018-06-15 18:04:31.401 JST [128004] LOG:  starting logical decoding
for slot "s1"
2018-06-15 18:04:31.401 JST [128004] DETAIL:  Streaming transactions
committing after 0/2000000, reading WAL from 0/1645388.
2018-06-15 18:04:31.401 JST [128004] STATEMENT:  select
pg_replication_slot_advance('s1', pg_current_wal_lsn());
2018-06-15 18:04:31.402 JST [128004] LOG:  logical decoding found
consistent point at 0/1645388
2018-06-15 18:04:31.402 JST [128004] DETAIL:  There are no running transactions.
2018-06-15 18:04:31.402 JST [128004] STATEMENT:  select
pg_replication_slot_advance('s1', pg_current_wal_lsn());
 pg_replication_slot_advance
-----------------------------
 (s1,0/2017828)
(1 row)

[1] https://www.postgresql.org/message-id/34d66f63-40a9-4c3e-c9a1-248d1e393d29%40enterprisedb.com

Regards,

--
Masahiko Sawada
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center


В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tatsuo Ishii
Дата:
Сообщение: Re: Memory leaks in BufFileOpenShared()
Следующее
От: Amit Langote
Дата:
Сообщение: Re: why partition pruning doesn't work?