[pgpool-general: 415] Re: pgpool2 status info

[Lou Kamenov]

Keeping large set of wal segments on the master, could be avoided by
using shipping to another server, amazon s3, etc.
There are tools like WAL-E that handle this seamlessly.

On Mon, May 7, 2012 at 5:04 AM, Benjamin Fras <benjamin.fras at petafuel.de> wrote:
> Hi,
>
> I can confirm your assumptions. We've combined two approaches in our environment:
> A) Increasing the value of wal_keep_segments for a "small" downtime of the slave nodes
> B) Implementing a full base backup routine using online recovery and the recovery_1st_stage_command
>
> We are using pgpooladmin and the online recovery does a full base backup. If our Slave goes down just for a small period of time, all we need to do to bring it up again, is just starting the postgres service. I have to mention that we use the native postgres streaming replication and pgpool is used for high-availability purposes only.
>
> Regarding the disconnect function in pgpooladmin; I've also experienced that disconnecting the master node does not trigger a failover and your cluster remains without a master. As a result you can't write to the db anymore, as the slave is write-protected and read-only. I don't see the sense and I assume that this might be a bug.
>
> PS: I hope I can provide you the promised documentation by the end of this week.
>
> Regards,
>
> petaFuel GmbH
> Benjamin Fras
> IT Systemadministration
>
> Tel.: +49 (0) 8161 4060 - 222
>
> Geschäftsführer: Ludwig Adam, Dr. Peter Schönweitz
> Amtsgericht München, HRB 133773
>
>
> -----Ursprüngliche Nachricht-----
> Von: pgpool-general-bounces at pgpool.net [mailto:pgpool-general-bounces at pgpool.net] Im Auftrag von Tatsuo Ishii
> Gesendet: Sonntag, 6. Mai 2012 03:55
> An: videanuadrian at yahoo.com
> Cc: pgpool-general at pgpool.net
> Betreff: [pgpool-general: 413] Re: pgpool2 status info
>
>> Hi Ruben,
>> I have modified the wal_keep_segments to 5000, but i guess that this should be parametrized depending on how long do you plan to keep the slave down and how much traffic do you have towards the database. Thanks for the hint.
>>
>> I`m planning to have a slave node on a server the is powered off every night, so just before the machine will shutdown i will detach the node from cluster and then stop postgresql process. Then in the morning i will start the postgresql process and after 1 min i will attach the node to the cluster.
>>
>> Regarding the second question maybe we can find someone to confirm our suppositions :)
>
> I think Ruben is right.
> --
> Tatsuo Ishii
> SRA OSS, Inc. Japan
> English: http://www.sraoss.co.jp/index_en.php
> Japanese: http://www.sraoss.co.jp
> _______________________________________________
> pgpool-general mailing list
> pgpool-general at pgpool.net
> http://www.pgpool.net/mailman/listinfo/pgpool-general
> _______________________________________________
> pgpool-general mailing list
> pgpool-general at pgpool.net
> http://www.pgpool.net/mailman/listinfo/pgpool-general