W dniu 2010-09-29 10:56, Guillaume Lelarge pisze:
>
> Using rar on the data folder should work if you don't have tablespaces.
> If you have some, add them to your data rar archive.
>
> And instead of overwriting the data folder on the slave, wipe out your
> old data folder and recreate it by restoring your rar archive.
>
> BTW, pgAdmin backup and restore facilities use pg_dump/pg_restore. So
> it's not really surprising that you get the same results :)
>
>
Alright, now it starts - thanks:
2010-09-29 14:54:33 CEST LOG: database system was interrupted; last
known up at 2010-09-29 13:02:11 CEST
2010-09-29 14:54:33 CEST LOG: entering standby mode
2010-09-29 14:54:34 CEST LOG: streaming replication successfully
connected to primary
2010-09-29 14:54:34 CEST LOG: redo starts at 6/D2000020
2010-09-29 14:54:34 CEST LOG: consistent recovery state reached at
6/D3000000
2010-09-29 14:54:34 CEST LOG: database system is ready to accept read
only connections
, but still some questions on my side :)
1. I wiped out the data dir on standby and modified configuration in
files backed up from master - I believe I had to it, to configure it as
standby.
2. Now I would like to connect to standby to see myself if it
replicates, but when I run pgadminIII and try to connect to standby I got:
2010-09-29 14:55:05 CEST FATAL: could not open file "global/11601": No
such file or directory
And I can't connect to standby database. I know it's because I wiped out
/data folder - how can I connect to standby then? Even if I can believe
it's replicating at the moment, when the master is down then I need
access to standby.
Regards,
Michal