[pgpool-general: 4930] Re: Pgpool and IP bonding

[Avi Weinberg]

Thanks for the fast reply to my previous IP bonding issue.

I have a setup with two pgpools servers with pgpool failover capability enabled.
I do not know if this is also related to IP bonding, but when I write a shell script to constantly perform pgpool restart I very fast get the following "pgpool dead but subsys locked".  If I run the restart scrip on two terminals I get it in matter of seconds.


[root at postgres_server1 ~]# service pgpool status
pgpool dead but subsys locked


2016-08-18 15:27:02: pid 61622: LOG:  watchdog child process with pid: 61624 exits with status 768
2016-08-18 15:27:02: pid 61622: FATAL:  watchdog child process exit with fatal error. exiting pgpool-II
2016-08-18 15:27:02: pid 61625: LOG:  setting the local watchdog node name to "Linux_postgres_server1_9999"
2016-08-18 15:27:02: pid 61625: LOG:  watchdog cluster configured with 1 remote nodes
2016-08-18 15:27:02: pid 61625: LOG:  watchdog remote node:0 on 172.18.255.42:9000
2016-08-18 15:27:02: pid 61625: LOG:  interface monitoring is disabled in watchdog
2016-08-18 15:27:04: pid 61644: FATAL:  could not read pid file
2016-08-18 15:27:04: pid 61685: LOG:  reading status file: 1 th backend is set to down status
2016-08-18 15:27:04: pid 61685: LOG:  waiting for watchdog to initialize
2016-08-18 15:27:04: pid 61687: LOG:  setting the local watchdog node name to "Linux_postgres_server1_9999"
2016-08-18 15:27:04: pid 61687: LOG:  watchdog cluster configured with 1 remote nodes
2016-08-18 15:27:04: pid 61687: LOG:  watchdog remote node:0 on 172.18.255.42:9000
2016-08-18 15:27:04: pid 61687: LOG:  interface monitoring is disabled in watchdog
2016-08-18 15:27:04: pid 61687: FATAL:  failed to create watchdog receive socket
2016-08-18 15:27:04: pid 61687: DETAIL:  bind on "0.0.0.0:cslistener" failed with reason: "Address already in use"
2016-08-18 15:27:04: pid 61685: LOG:  watchdog child process with pid: 61687 exits with status 768
2016-08-18 15:27:04: pid 61685: FATAL:  watchdog child process exit with fatal error. exiting pgpool-II
2016-08-18 15:27:04: pid 61688: LOG:  setting the local watchdog node name to "Linux_postgres_server1_9999"
2016-08-18 15:27:04: pid 61688: LOG:  watchdog cluster configured with 1 remote nodes
2016-08-18 15:27:04: pid 61688: LOG:  watchdog remote node:0 on 172.18.255.42:9000
2016-08-18 15:27:04: pid 61688: LOG:  interface monitoring is disabled in watchdog
2016-08-18 15:27:05: pid 61708: FATAL:  could not read pid file
2016-08-18 15:27:05: pid 61748: LOG:  reading status file: 1 th backend is set to down status
2016-08-18 15:27:05: pid 61748: LOG:  waiting for watchdog to initialize
2016-08-18 15:27:05: pid 61750: LOG:  setting the local watchdog node name to "Linux_postgres_server1_9999"
2016-08-18 15:27:05: pid 61750: LOG:  watchdog cluster configured with 1 remote nodes
2016-08-18 15:27:05: pid 61750: LOG:  watchdog remote node:0 on 172.18.255.42:9000
2016-08-18 15:27:05: pid 61750: LOG:  interface monitoring is disabled in watchdog
2016-08-18 15:27:05: pid 61750: FATAL:  failed to create watchdog receive socket
2016-08-18 15:27:05: pid 61750: DETAIL:  bind on "0.0.0.0:cslistener" failed with reason: "Address already in use"
2016-08-18 15:27:05: pid 61748: LOG:  watchdog child process with pid: 61750 exits with status 768
2016-08-18 15:27:05: pid 61748: FATAL:  watchdog child process exit with fatal error. exiting pgpool-II
2016-08-18 15:27:05: pid 61751: LOG:  setting the local watchdog node name to "Linux_postgres_server1_9999"
2016-08-18 15:27:05: pid 61751: LOG:  watchdog cluster configured with 1 remote nodes
2016-08-18 15:27:05: pid 61751: LOG:  watchdog remote node:0 on 172.18.255.42:9000
2016-08-18 15:27:05: pid 61751: LOG:  interface monitoring is disabled in watchdog
2016-08-18 15:27:06: pid 61770: FATAL:  could not read pid file
2016-08-18 15:27:07: pid 61810: LOG:  reading status file: 1 th backend is set to down status
2016-08-18 15:27:07: pid 61810: LOG:  waiting for watchdog to initialize
2016-08-18 15:27:07: pid 61812: LOG:  setting the local watchdog node name to "Linux_postgres_server1_9999"
2016-08-18 15:27:07: pid 61812: LOG:  watchdog cluster configured with 1 remote nodes
2016-08-18 15:27:07: pid 61812: LOG:  watchdog remote node:0 on 172.18.255.42:9000
2016-08-18 15:27:07: pid 61812: LOG:  interface monitoring is disabled in watchdog
2016-08-18 15:27:07: pid 61812: FATAL:  failed to create watchdog receive socket
2016-08-18 15:27:07: pid 61812: DETAIL:  bind on "0.0.0.0:cslistener" failed with reason: "Address already in use"
2016-08-18 15:27:07: pid 61810: LOG:  watchdog child process with pid: 61812 exits with status 768
2016-08-18 15:27:07: pid 61810: FATAL:  watchdog child process exit with fatal error. exiting pgpool-II
2016-08-18 15:27:07: pid 61813: LOG:  setting the local watchdog node name to "Linux_postgres_server1_9999"
2016-08-18 15:27:07: pid 61813: LOG:  watchdog cluster configured with 1 remote nodes
2016-08-18 15:27:07: pid 61813: LOG:  watchdog remote node:0 on 172.18.255.42:9000
2016-08-18 15:27:07: pid 61813: LOG:  interface monitoring is disabled in watchdog
2016-08-18 15:27:35: pid 58439: LOG:  watchdog: lifecheck started
^C
IMPORTANT - This email and any attachments is intended for the above named addressee(s), and may contain information which is confidential or privileged. If you are not the intended recipient, please inform the sender immediately and delete this email: you should not copy or use this e-mail for any purpose nor disclose its contents to any person.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20160818/61f39f6d/attachment.html>