[pgpool-hackers: 3777] watchdog child process exits (issue625)

[Bo Peng]

Hi Usama,

While working on the following issue, I found that
watchdog child process exits without restart
when "Network is unreachable" error occurs.
And systemd terminates the pgpool main process.

  https://www.pgpool.net/mantisbt/view.php?id=625

Sorry, I am not familiar with watchdog behavior.
Is it the correct behavior?
Watchdog child process should not be waiting for the network to come back?

BTW, I couldn't reproduce the [IN NETWORK TROUBLE] state.

========== pgpool.log ==========
WARNING:  network IP is removed and system has no IP is assigned           
DETAIL:  changing the state to in network trouble
LOG:  watchdog node state changed from [STANDBY] to [IN NETWORK TROUBLE]
FATAL:  system has lost the network
LOG:  Watchdog is shutting down
LOG:  watchdog child process with pid: 228427 exits with status 768 
FATAL:  watchdog child process exit with fatal error. exiting pgpool-II
LOG:  setting the local watchdog node name to "192.168.164.202:9999 Linux postmetahive2"
LOG:  watchdog cluster is configured with 2 remote nodes
LOG:  watchdog remote node:0 on 192.168.164.201:9000
LOG:  watchdog remote node:1 on 192.168.164.203:9000
LOG:  interface monitoring is disabled in watchdog
LOG:  connect on socket failed
DETAIL:  connect failed with error: "Network is unreachable"
LOG:  connect on socket failed
DETAIL:  connect failed with error: "Network is unreachable"
LOG:  watchdog node state changed from [DEAD] to [LOADING]
LOG:  Watchdog is shutting down
==============================

Regards,
-- 
Bo Peng <pengbo at sraoss.co.jp>
SRA OSS, Inc. Japan