postgres-xc-general Mailing List for Postgres-XC

Hi Koichi,Masataka

Successfully added *gtm* proxy but having issue with coordinator and
datanode.

*Coordinator :*
add coordinator master name host port pooler dir
here what value i should use in place of pooler ? pooler port ?

pgxc_ctl add coordinator master coord3 node5 20004 20010
/home/postgres/pgxc/nodes/coord
ERROR: sorry found some inconflicts in coordinator master configuration.

pgxc_ctl add coordinator master coord3 node5 20004
/home/postgres/pgxc/nodes/coord
ERROR: please specify the working director for the coordinator master

*Datanode :*
add datanode master name host port dir
pgxc_ctl add datanode master new node5 20008
/home/postgres/pgxc/nodes/dn_master
ERROR: sorry found some inconflicts in datanode master configuration.

Does earlier manual installation could be reason for this issue ?

*P.S:*
*gtm_prxy* : Nodename: 'gtm_pxy3', port: 20001, dir:
'/home/postgres/pgxc/nodes/gtm_pxy'
*Coordinator Master*: Nodename: 'coord3', port: 20004, pooler port: 20010,
dir:' /home/postgres/pgxc/nodes/coord'
*Datanode Master* :   Nodename: 'datanode3', port: 20008, dir:'
/home/postgres/pgxc/nodes/dn_master'

please look at my commands if its erroneous

Please advice.

Regards,
Juned Khan



On Wed, Mar 12, 2014 at 11:45 AM, Juned Khan <jkh...@gm...> wrote:

> I tried to add coordinator and datanode manually
> and now it seems everything is messed up.
>
> Every component is just stopped working, if i kill all and
> start all again then again its shows everything is running
> but in that case it takes too long to connect dtabase(Psql -dkoichi)
>
> and at last it just fails to connect to the database
>
>
>
>
>
>
> *PGXC Psql -dkoichiSelected coord1.psql: server closed the connection
> unexpectedly         This probably means the server terminated
> abnormally        before or while processing the request.PGXC *
>
> Please advice.
>
>
>
> On Wed, Mar 12, 2014 at 4:34 AM, Michael Paquier <
> mic...@gm...> wrote:
>
>> On Wed, Mar 12, 2014 at 4:10 AM, Masataka Saito <pg...@gm...> wrote:
>> > On 11 March 2014 16:42, Juned Khan <jkh...@gm...> wrote:
>> >> Hi Masataka,
>> >>
>> >> For testing i am tried to get this work added manual query for CREATE
>> NODE.
>> >>
>> >> Now its look like this:
>> >>
>> >> postgres=# \q
>> >> PGXC Psql -dpostgres -h node5 -p 20004
>> >> Selected coord3.
>> >>
>> >> postgres=# select * from pgxc_node;
>> >>  node_name | node_type | node_port | node_host | nodeis_primary |
>> >> nodeis_preferred |   node_id
>> >>
>> -----------+-----------+-----------+-----------+----------------+------------------+-------------
>> >>  coord3    | C         |      5432 | localhost | f              | f
>> >> |  1638403545
>> >>  coord1    | C         |     20004 | node3     | f              | f
>> >> |  1885696643
>> >>  coord2    | C         |     20005 | node4     | f              | f
>> >> | -1197102633
>> >>  datanode1 | D         |     20008 | node3     | f              | f
>> >> |   888802358
>> >>  datanode2 | D         |     20009 | node4     | f              | f
>> >> |  -905831925
>> >>  datanode3 | D         |     20008 | node5     | f              | f
>> >> | -1894792127
>> >> (6 rows)
>> >>
>> >> Here coord3 runs on 5432 which should 20004., it creates problem i
>> think
>> >> isn't it ?
>> >>
>> >
>> > As far as I know, the local node information is not important and it
>> > is not necessary to update.
>> > "pgxc_ctl add" corrects it by "ALTER NODE".
>> CREATE NODE and DROP NODE only run locally. Why not ALTER NODE for
>> consistency? Am I missing smth?
>> --
>> Michael
>>
>
>
>
> --
> Thanks,
> Juned Khan
> iNextrix Technologies Pvt Ltd.
> www.inextrix.com
>



-- 
Thanks,
Juned Khan
iNextrix Technologies Pvt Ltd.
www.inextrix.com

I tried to add coordinator and datanode manually
and now it seems everything is messed up.

Every component is just stopped working, if i kill all and
start all again then again its shows everything is running
but in that case it takes too long to connect dtabase(Psql -dkoichi)

and at last it just fails to connect to the database






*PGXC Psql -dkoichiSelected coord1.psql: server closed the connection
unexpectedly        This probably means the server terminated
abnormally        before or while processing the request.PGXC *

Please advice.



On Wed, Mar 12, 2014 at 4:34 AM, Michael Paquier
<mic...@gm...>wrote:

> On Wed, Mar 12, 2014 at 4:10 AM, Masataka Saito <pg...@gm...> wrote:
> > On 11 March 2014 16:42, Juned Khan <jkh...@gm...> wrote:
> >> Hi Masataka,
> >>
> >> For testing i am tried to get this work added manual query for CREATE
> NODE.
> >>
> >> Now its look like this:
> >>
> >> postgres=# \q
> >> PGXC Psql -dpostgres -h node5 -p 20004
> >> Selected coord3.
> >>
> >> postgres=# select * from pgxc_node;
> >>  node_name | node_type | node_port | node_host | nodeis_primary |
> >> nodeis_preferred |   node_id
> >>
> -----------+-----------+-----------+-----------+----------------+------------------+-------------
> >>  coord3    | C         |      5432 | localhost | f              | f
> >> |  1638403545
> >>  coord1    | C         |     20004 | node3     | f              | f
> >> |  1885696643
> >>  coord2    | C         |     20005 | node4     | f              | f
> >> | -1197102633
> >>  datanode1 | D         |     20008 | node3     | f              | f
> >> |   888802358
> >>  datanode2 | D         |     20009 | node4     | f              | f
> >> |  -905831925
> >>  datanode3 | D         |     20008 | node5     | f              | f
> >> | -1894792127
> >> (6 rows)
> >>
> >> Here coord3 runs on 5432 which should 20004., it creates problem i think
> >> isn't it ?
> >>
> >
> > As far as I know, the local node information is not important and it
> > is not necessary to update.
> > "pgxc_ctl add" corrects it by "ALTER NODE".
> CREATE NODE and DROP NODE only run locally. Why not ALTER NODE for
> consistency? Am I missing smth?
> --
> Michael
>



-- 
Thanks,
Juned Khan
iNextrix Technologies Pvt Ltd.
www.inextrix.com