[pgpool-general: 921] recovery command failed but SELECT pgpool_recovery(...) works

[Kent Tong]

Hi,

I am using 3.1.3 and working on the online recovery. After issuing the
"pcp_recovery_node -d 10 localhost 9898 postgres postgres 0" command to try
to recover backend 0, it fails:

EBUG: send: tos="R", len=46
DEBUG: recv: tos="r", len=21, data=AuthenticationOK
DEBUG: send: tos="D", len=6
DEBUG: recv: tos="e", len=20, data=recovery failed
DEBUG: command failed. reason=recovery failed
BackendError
DEBUG: send: tos="X", len=4

The following messages are found in the log. It means that pgpool connected
the to backend 0 successfully but the SELECT failed:

Aug 28 12:32:44 pgpool02 pgpool[3791]: starting recovering node 0

Aug 28 12:32:44 pgpool02 pgpool[3791]: starting recovery command: "SELECT
pgpool_recovery('basebackup.sh', 'masterdb', '/var/lib/pgsql/9.1/data')"

Aug 28 12:32:44 pgpool02 pgpool[3791]: exec_recovery: basebackup.sh command
failed at 1st stage

However, I can use psql to execute that SELECT query on the pgpool machine
successfully:

[root at pgpool02 pgpool-II-91]# psql -h masterdb -c "SELECT
pgpool_recovery('basebackup.sh', 'masterdb', '/var/lib/pgsql/9.1/data')"
template1 postgres
Password for user postgres:

 pgpool_recovery

-----------------

 t

(1 row)

The recovery_user and recovery_password have been set to "postgres" and its
password.

So, any idea why it failed?

-- 
Kent Tong
IT author and consultant, child educator
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20120828/691c9ecd/attachment.html>