[pgpool-general: 3290] Re: Strange behaviour of pcp_watchdog_info

[Karthick Thanigaimani]

yes thats correct.


________________________________
 From: Yugo Nagata <nagata at sraoss.co.jp>
To: Karthick Thanigaimani <kargold100 at yahoo.com> 
Cc: "pgpool-general at pgpool.net" <pgpool-general at pgpool.net> 
Sent: Monday, 17 November 2014 6:35 PM
Subject: Re: [pgpool-general: 3284] Strange behaviour of pcp_watchdog_info
 

Hi,

On Fri, 14 Nov 2014 19:35:53 +0800
Karthick Thanigaimani <kargold100 at yahoo.com> wrote:
> These are run the the same servers and we checked the _otherhostname_ it looks fine only.

You mean, the same two pcp_command were run at a server, but the results
were different; the first one returned BackendError and the second succeeded?


On Fri, 14 Nov 2014 19:50:51 +0800
Karthick Thanigaimani <kargold100 at yahoo.com> wrote:
> These are the files and values. Let me know if aything elseto be added?
> 
> more pgpool.conf | grep pgpool_hostname
> other_pgpool_hostname0 = '10.0.90.11'
> more pgpool.conf_10.0.90.11 | grep pgpool_hostname
> other_pgpool_hostname0 = '10.0.90.12'
> more pgpool.conf_10.0.90.12 | grep pgpool_hostname
> other_pgpool_hostname0 = '10.0.90.11'

I assume there are two machine (IP: 10.0.90.{11,12}) each of which has
pgpool-II installed, and the conf file is pgpool.conf_${IP}. 

It seems no problem. If you want to know watchdog status of other pgpool
with this pgpool.conf, the command to be run is; 

$ pcp_watchdog_info 0 node1 9898 postgres test 0

since the last argument is the postfix number of "ohter_pgpool_hostnameX".

So, when the last argument is specified to 2 this returns BackendError
since other_pgpool_hostname2 isn't specified.

> 
> 
> 
> ________________________________
>  From: Yugo Nagata <nagata at sraoss.co.jp>
> To: Karthick Thanigaimani <kargold100 at yahoo.com> 
> Cc: "pgpool-general at pgpool.net" <pgpool-general at pgpool.net> 
> Sent: Friday, 14 November 2014 7:08 PM
> Subject: Re: [pgpool-general: 3284] Strange behaviour of pcp_watchdog_info
>  
> 
> Hi,
> 
> On Fri, 14 Nov 2014 08:40:20 +0800
> Karthick Thanigaimani <kargold100 at yahoo.com> wrote:
> 
> > Hi,
> > 
> > We were facing a Backenderror when we try to check the status of the node using the pcp_watchdog_info command.
> > we have postgres 9.3.4 and pgpool 3.3.3 .
> > 
> > $ psql -p9999 -h node1 -c "show pool_nodes"
> > Password:
> >  node_id |  hostname  | port | status | lb_weight |  role
> > ---------+------------+------+--------+-----------+---------
> >  0       | node1 | 5432 | 2      | 0.500000  | primary
> >  1       | node2 | 5432 | 2      | 0.500000  | standby
> > (2 rows)
> > 
> > $ pcp_watchdog_info 10 node1 9898 postgres test 1
> > BackendError
> > 
> > $ pcp_watchdog_info 10 node1 9898 postgres test 1
> > node2 9999 9000 2
> 
> Are these tow results run in deffernt servers? If so, please
> check other_pgpool_hostnameX of pgpool.conf in the both servers.
> One of them may not have specification of ohter_pgopol_hostname2.
> 
> 
> > 
> > the nodes are in sync and the DB / pgpool is online as well.
> > 
> > can you please advise if this is a bug or what could be the reason for this.
> > 
> > regards
> > Karthick



> > 
> 
> 
> -- 
> Yugo Nagata <nagata at sraoss.co.jp
> >


-- 
Yugo Nagata <nagata at sraoss.co.jp>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.pgpool.net/pipermail/pgpool-general/attachments/20141118/e2659d2c/attachment.htm>