[Pgpool-general] Error message when sending query that causes failover

[Jaume Sabater]

Tatsuo Ishii wrote:
 
>> Aha, it makes sense. So the desired and expected behaviour is the
>> one I am having when the slave node fails (query becoming the way
>> pgpool-II realises a node is down, doing failover of that node, and
>> executing the actual query on the rest of nodes). Correct?
> 
> In this case you need to reconnect to pgpool-II since it disconnects 
> all the connections from clients to pgpool-II.

Including the one from the query that caused pgpool-II to realise that
node was not reachable? Does not seem so as the query was apparently
executed correctly.

Given my system has web front-ends that create connections every time,
it is not a problem for me. It's just fine.

-- 
Jaume Sabater
http://linuxsilo.net/

"Ubi sapientas ibi libertas"