[pgpool-general: 1331] pgpool watchdog setting

[ning chan]

hi all,
I am trying to setup watchdog for my 2 nodes cluster (primary - standby,
and will add another node later).
After i configure everything on my primary node (while pgpool on the other
servers are down), I see the following in the syslog:

Jan 20 16:15:30 se032c-94-30 pgpool[29047]: wd_chk_sticky: ifup[/sbin/ip]
doesn't have sticky bit
Jan 20 16:15:30 se032c-94-30 pgpool[29048]: watchdog: wd_check_config failed
Jan 20 16:15:30 se032c-94-30 pgpool[29048]: wd_main error
Jan 20 16:15:30 se032c-94-30 pgpool[29048]: unlink(/tmp/.s.PGSQL.9898)
failed: No such file or directory

I google a bit, looks like sticky bit is not set to /sbin folder by
default, should i make change to system folder? why it's matter with pgpool?

here is my watchdog setting in pgpool.conf

#------------------------------------------------------------------------------
# WATCHDOG
#------------------------------------------------------------------------------
use_watchdog = on
trusted_servers = '10.89.94.31'
delegate_IP = '10.89.94.13'
wd_hostname = 'se032c-94-30'
wd_port = 9000
wd_interval = 10
ping_path = '/bin'
ifconfig_path = '/sbin'

if_up_cmd = 'ip addr add $_IP_$/32 dev eth0'     < -- instead of using ifup
and ifdown, i use the ip addr to bring up the virtual IP. I tried the
eth0:0 but it did not work for me, but the ip addr add/del works just fine

if_down_cmd = 'ip addr del $_IP_$/32 dev eth0'
arping_path = '/sbin'           # arping command path
arping_cmd = 'arping -U $_IP_$ -w 1'
wd_life_point = 3
wd_lifecheck_query = 'SELECT 1'
#other_pgpool_hostname0 = 'host0'
#other_pgpool_port0 = 5432
#other_wd_port0 = 9000

Can someone give me some help here?

Thanks~
Ning
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20130120/93017bb8/attachment.html>