[Pgcluster-general] S.O.S., replication sever failover

["Tomás A. Rossi"]

Marcelo/Porks escribió:
> On 1/8/07, "Tomás A. Rossi" <tomas at mecon.gov.ar> wrote:
>   
>> But isn't that the whole idea? I want two independent nodes, if one of
>> them fails, I want the other one to take over. If "ldapserver" fails,
>> which is what I'm trying to simulate, "ldapserver2" has to take the
>> post. Don't wanna have single fail-points.
>>     
>
> Yes... but at your configuration of ldapserver2 you set:
> <When_Stand_Alone>    read_only </When_Stand_Alone>
>
> So when the ldapserver goes down... the ldapserver2 is alone... and...
> it's read_only...
>
> But... try to do what Mrs Mitani wrote. He is the master.... I'm only
> a padawan :D
>
>   
I see your point but in the config template it says:

#       o When_Stand_Alone : When all replication servers fell,
#                            you can set up two kinds of permittion,
#                            "real_only" or "read_write".

... and in this case, not all replication server fell but one remains 
operative. Either way, I'm not receiving an error message like if I was 
trying to write over a read_only DB. It just hangs. :(

Don't worry, I'm going to collect some logs and reply to Lord Mitani ;)

Regards,
--
Tom;
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://pgfoundry.org/pipermail/pgcluster-general/attachments/20070108/2789607a/attachment.html