Re: [HACKERS] the include-timestamp data returned by pg_logical_slot_peek_changes() is always 2000-01-01 in 9.5.1

2016-03-09 Thread hailong . li
On 2016年03月09日 23:31, Andres Freund wrote: Did you enable track_commit_timestamps in the server? That's unrelated, commit ts is about an xid timestamp mapping. Yes, whether enable track_commit_timestamp or not, I have just done the test again and the result is the same. Thxs all!

Re: [HACKERS] the include-timestamp data returned by pg_logical_slot_peek_changes() is always 2000-01-01 in 9.5.1

2016-03-09 Thread Andres Freund
On March 9, 2016 4:26:01 AM PST, Craig Ringer wrote: >On 9 March 2016 at 18:13, 李海龙 wrote: > >> >> >> HI, pgsql-hackers >> >> The include-timestamp data returned by >pg_logical_slot_peek_changes() is >> always 2000-01-01 in 9.5.1, is it not normal？

Re: [HACKERS] the include-timestamp data returned by pg_logical_slot_peek_changes() is always 2000-01-01 in 9.5.1

2016-03-09 Thread Craig Ringer
On 9 March 2016 at 18:13, 李海龙 wrote: > > > HI, pgsql-hackers > > The include-timestamp data returned by pg_logical_slot_peek_changes() is > always 2000-01-01 in 9.5.1, is it not normal？ > Did you enable track_commit_timestamps in the server? If not, and it's returning

[HACKERS] the include-timestamp data returned by pg_logical_slot_peek_changes() is always 2000-01-01 in 9.5.1

2016-03-09 Thread 李海龙
HI, pgsql-hackers The include-timestamp data returned by pg_logical_slot_peek_changes() is always 2000-01-01 in 9.5.1, is it not normal？ The following is the test: [postgres@pgtest ~]$ cat /etc/issue CentOS release 6.4 (Final) Kernel \r on an \m [postgres@pgtest ~]$ uname -av Linux