Re: Reviving lost replication slots - Mailing list pgsql-hackers

From sirisha chamarthi
Subject Re: Reviving lost replication slots
Date
Msg-id CAKrAKeXwUOTEVNtw=Jt6ZmFSJsNcFrYvtOhy0oJj_tGefLVB3A@mail.gmail.com
Whole thread Raw
In response to Re: Reviving lost replication slots  (Amit Kapila <amit.kapila16@gmail.com>)
Responses Re: Reviving lost replication slots
List pgsql-hackers


On Wed, Nov 9, 2022 at 2:37 AM Amit Kapila <amit.kapila16@gmail.com> wrote:
On Fri, Nov 4, 2022 at 1:40 PM sirisha chamarthi
<sirichamarthi22@gmail.com> wrote:
>
 Is the intent of setting restart_lsn to InvalidXLogRecPtr was to
disallow reviving the slot?
>

I think the intent is to compute the correct value for
replicationSlotMinLSN as we use restart_lsn for it and using the
invalidated slot's restart_lsn value for it doesn't make sense.

 Correct. If a slot is invalidated (lost), then shouldn't we ignore the slot from computing the catalog_xmin?  I don't see it being set to InvalidTransactionId in ReplicationSlotsComputeRequiredXmin. Attached a small patch to address this and the output after the patch is as shown below.

postgres=# select * from pg_replication_slots;
 slot_name |    plugin     | slot_type | datoid | database | temporary | active | active_pid | xmin | catalog_xmin | restart_lsn | confirmed_flush_lsn | wal_status | safe_wal_size | two_phase
-----------+---------------+-----------+--------+----------+-----------+--------+------------+------+--------------+-------------+---------------------+------------+---------------+-----------
 s2        | test_decoding | logical   |      5 | postgres | f         | f      |            |      |          771 | 0/30466368  | 0/304663A0          | reserved   |      28903824 | f
(1 row)

postgres=# create table t2(c int, c1 char(100));
CREATE TABLE
postgres=# drop table t2;
DROP TABLE
postgres=# vacuum pg_class;
VACUUM
postgres=# select n_dead_tup from pg_stat_all_tables where relname = 'pg_class';
 n_dead_tup
------------
          2
(1 row)

postgres=# select * from pg_stat_replication;
 pid | usesysid | usename | application_name | client_addr | client_hostname | client_port | backend_start | backend_xmin | state | sent_lsn | write_lsn | flush_lsn | replay_lsn | write_lag | flush_lag | replay_lag | sync_pri
ority | sync_state | reply_time
-----+----------+---------+------------------+-------------+-----------------+-------------+---------------+--------------+-------+----------+-----------+-----------+------------+-----------+-----------+------------+---------
------+------------+------------
(0 rows)

postgres=# insert into t1 select * from t1;
INSERT 0 2097152
postgres=# checkpoint;
CHECKPOINT
postgres=# select * from pg_replication_slots;
 slot_name |    plugin     | slot_type | datoid | database | temporary | active | active_pid | xmin | catalog_xmin | restart_lsn | confirmed_flush_lsn | wal_status | safe_wal_size | two_phase
-----------+---------------+-----------+--------+----------+-----------+--------+------------+------+--------------+-------------+---------------------+------------+---------------+-----------
 s2        | test_decoding | logical   |      5 | postgres | f         | f      |            |      |          771 |             | 0/304663A0          | lost       |               | f
(1 row)

postgres=# vacuum pg_class;
VACUUM
postgres=# select n_dead_tup from pg_stat_all_tables where relname = 'pg_class';
 n_dead_tup
------------
          0
(1 row)
 

--
With Regards,
Amit Kapila.

pgsql-hackers by date:

Previous
From: Amit Kapila
Date:
Subject: Assertion failure in SnapBuildInitialSnapshot()
Next
From: sirisha chamarthi
Date:
Subject: Re: Reviving lost replication slots