[pgpool-general: 5026] Re: Failover command not invoked by secondary pgpool after its promotion to primary in HA

[Tatsuo Ishii]

That makes sense.

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp

> Hi Tatsuo,
> 
> I guess higher default value could be better, because there is less drawback to have a high value than a low value for this parameter.
> 
> Regards,
> 
> Christophe
> 
> -----Message d'origine-----
> De : Tatsuo Ishii [mailto:ishii at sraoss.co.jp] 
> Envoyé : vendredi 23 septembre 2016 10:11
> À : Christophe Le Roux <christophe.le.roux at fr.clara.net>
> Cc : monfardini at ldpgis.it; pgpool-general at pgpool.net
> Objet : Re: [pgpool-general: 5023] Re: Failover command not invoked by secondary pgpool after its promotion to primary in HA
> 
> Christophe,
> 
> Thank you for studying it. Do you think it's a good idea to change the default value of search_primary_node_timeout to say, 300 in the pgpool.conf sample files?
> 
> Best regards,
> --
> Tatsuo Ishii
> SRA OSS, Inc. Japan
> English: http://www.sraoss.co.jp/index_en.php
> Japanese:http://www.sraoss.co.jp
> 
>> Hi Gabriele,
>> 
>> If your slave make more than 10 seconds to be elected as primary (which can be dependant of other timeout parameters), PGpool will never find it ( PGpool stop to search for a primary node when the search_primary_node_timeout is reached).
>> 10 seconds is very short, try with a higher value (60 or 120). 300 is a value that I found in a blog who explain a 2 pgpool and 2 Postgresql setup.
>> 
>> Regards,
>> 
>> Christophe
>> 
>> De : gabrimonfa at gmail.com [mailto:gabrimonfa at gmail.com] De la part de 
>> Gabriele Monfardini Envoyé : jeudi 22 septembre 2016 18:16 À : 
>> Christophe Le Roux <christophe.le.roux at fr.clara.net> Cc : 
>> pgpool-general at pgpool.net Objet : Re: [pgpool-general: 5010] Re: 
>> Failover command not invoked by secondary pgpool after its promotion 
>> to primary in HA
>> 
>> Hi Christophe, hi all,
>> 
>> 
>> On Wed, Sep 21, 2016 at 6:10 PM, Christophe Le Roux <christophe.le.roux at fr.clara.net<mailto:christophe.le.roux at fr.clara.net>> wrote:
>> I had the same issue when my search_primary_node_timeout had a bad value. With a value at 300, it work fine for me.
>> 
>> 
>> in our setup we have not modified default value for search_primary_node_timeout, which is 10.
>> I don't think it is needed to set to such an high value, 300.
>> Why do you think this value is involved?
>> 
>> Best regards,
>> 
>> Gabriele Monfardini
>> 
>> -----
>> Gabriele Monfardini
>> LdP Progetti GIS
>> tel: 0577.531049
>> email: monfardini at ldpgis.it<mailto:monfardini at ldpgis.it>
>>