2017-01-12 13:23 GMT+13:00 Adrian Klaver <adrian.klaver@xxxxxxxxxxx>:
On 01/11/2017 04:08 PM, Patrick B wrote:
Hi guys,
I'm using PostgreSQL 9.2 in two different servers.
server1 (Master Postgres DB server, running Postgres 9.2 / 128GB ram) -
RAID 10 Magnetic disks
server2 (Master Postgres DB server, running Postgres 9.2 / 128GB ram) -
EBS (AWS) io2 10k IOPS
When I run a query, I get this error:
ERROR: canceling statement due to statement timeout
statement_timeout is 0 in both servers.
However, on server1 I am able to run the query. Only on server2 that I
get that error.
Why? If it is same DB???
It is not the same DB if it is on two different servers not connected by replication. More to the point statement_timeout is a client connection setting, so is the client you use to connect to server2 the same as the one you use for server1?
Is AWS being 'helpful' and setting a timeout?
Is there anything in the log before the ERROR shown above that indicates something is setting statement_timeout?
Same database, different database servers; server1 is the old Master server and I'm using it to compare.
It is not the client, because if I run the query manually using explain analyze i get the error:
live_db=> explain analyze
SELECT DISTINCT id0
FROM
(SELECT
[...]ERROR: canceling statement due to statement timeout
just a remind that on server1 works, but on server2 it doesn't.
