Hi,
I have just committed changes to enhance online recovery in streaming
replication mode. Pgpool-II had to restart child process to recognize
failback node. Now pgpool-II becomes a little bit smarter and does not
need to restart children. This means that client sessions started
before online recovery can continue while doing online recovery.
The trick is to have local backend status cache in pgpool child. It
copies backend status into local memory. Thus the process continues to
disregard failback node until the session ends. Once the session ends,
the process exits with status 1, and pgpool-II main restart new one,
which will recognize failback node.
Note that this enhancement is applied to pcp_attach_node as well.
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp
