On 08/03/2016 07:21 PM, Patrick B wrote:
> Hi all,
>
> I'm currently using PostgreSQL 9.2. I noticed that the wal_files are
> being generated by the master well, no problems. But on the slaves, it
> seems to be a delay to the delivery of those wal_files.
>
> I got two slaves using streaming replication and wal files shipment from
> Master.
>
> *On the master:*
>
> ls -ltr /var/lib/pgsql/9.2/data/pg_xlogs/
> Aug 4 02:18 000000020000159D000000D1
> Aug 4 02:18 000000020000159D000000D2
> Aug 4 02:18 000000020000159D000000D3
>
>
> select * from pg_current_xlog_location();
> 159D/D6C8DAF8
>
>
> So, seems to be ok.
>
>
> *On the slave:*
>
>
> ls -ltr /var/lib/pgsql/9.2/wal_archive:
> Aug 4 00:58 000000020000159C00000071
> Aug 4 00:58 000000020000159C00000072
> Aug 4 00:58 000000020000159C00000073
>
>
> See the time difference? 2 hours? It seems the files are being delivered
> with 2 hours delay.
Both machines have same timezone?
How fast are you generating WALs?
How are you shipping the WALs?
>
> The streaming replication is working fine... But if that goes down, I'll
> need the wal_files up to date to recover the database.
>
> How can I see what's going on? What would be the steps? any tips?
>
> Cheers
> Patrick.
--
Adrian Klaver
adrian.klaver@aklaver.com