From: Koichi S. <koi...@gm...> - 2014-08-19 14:49:30
|
Thanks a lot for the report. It is pgxc_ctl bug. Replication should be synchronous. Will submit a fix soon. --- Koichi Suzuki 2014-08-19 23:40 GMT+09:00 zhou liang <zh...@gm...>: > Hi All, > Datanode streaming replication state is async after pgxc_ctl command > init all or start all。Datanode master sync_state is async in > pg_stat_replication,as fallows : > postgres=# select * from pg_stat_replication; > pid | usesysid | usename | application_name | client_addr | > client_hostname | client_port | backend_start | > state | sent_location | write_location | flush_location | > replay_location | sync_priority | sync_state > ------+----------+---------+------------------+----------------+-----------------+-------------+-------------------------------+-----------+---------------+----------------+----------------+-----------------+---------------+------------ > 2772 | 10 | postgres | datanode2 | 192.168.34.162 | > | 51715 | 2014-08-20 04:35:39.561691+08 | streaming > | 0/30011E8 | 0/30011E8 | 0/30011E8 | 0/30011B8 | > 0 | async > (1 row) > postgres=# show synchronous_standby_names; > synchronous_standby_names > --------------------------- > (1 row) > > behavior of pgxc_ctl : > 1、when stoping datanode slave,prepare_stopDatanodeSlave() set the > master to asynchronous mode by changing datanode master parameters > synchronous_standby_names='', then reload master postgresql.conf if > the master is running > 2、when starting datanode slave,prepare_startDatanodeSlave() change > master parameters synchronous_standby_names='datanode name' , but it > forget reload the master postgresql.conf, so the > synchronous_standby_names is invalid, replication is always > asynchronous, is it a bug of pgxc_ctl ? > > > -- > Thanks > Zhou Liang > > ------------------------------------------------------------------------------ > _______________________________________________ > Postgres-xc-bugs mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-bugs |