[pgpool-general: 8652] Re: online recovery causes node to loose replication state.

[Tatsuo Ishii]

> Hi Tatsuo,
> Please find the attached zip file containing the config files and logs you asked for.
> 
> Thank you in advance !

Thanks. Unfortunately there's no PostgreSQL logs. But I guess the reason for this:

>> My process to get the cluster back to normal.  Don't understand why the recovery process is failing...  Eventually, it works...
>> pcp_recovery_node -h 10.78.121.3 -p 9898 -U postgres -n 2
>> 
>> ERROR:  recovery is checking if postmaster is started
>> DETAIL:  postmaster on hostname:"catvmtspg04c.domain.net" 
>> database:"template1" user:"postgres" failed to start in 0 second
>> 
>> [postgres at catvmtspg04a pgpool-II]$ pcp_recovery_node -h 10.78.121.3 -p 
>> 9898 -U postgres -n 2
>> Password:
>> ERROR:  recovery is checking if postmaster is started
>> DETAIL:  postmaster on hostname:"catvmtspg04c.domain.net" 
>> database:"template1" user:"postgres" failed to start in 0 second

Because you set: recovery_timeout = 0. This will not give postmaster a
chance to spend sometime for internal processing before it starts
up. I'd suggest to get back to the default value (90).

Best reagards,
--
Tatsuo Ishii
SRA OSS LLC
English: http://www.sraoss.co.jp/index_en/
Japanese:http://www.sraoss.co.jp