postgres-xc-general Mailing List for Postgres-XC

Hi,

 I understand that the datanodes are read only and that updates/insert can
happen at co-ordinator. Also, it does not allow modification of column over
which the records are distributed.

However, in case I know what I am doing, it there anyway possible to modify
the values directly at datanodes. The modifications are not over column
over which distribution happens.

Thanks.
Sandeep

Dear Sir,  

My cluster has configured by pg_ctl now, but till filed.

Error message
--------------------------------------------------------------------------------
PGXC add coordinator master coord4 node4  20004 20010 /opt/pgxc/nodes/coord
…...  
Actual Command: ssh pgxc@node4 "( pg_ctl start -Z restoremode -D /opt/pgxc/nodes/coord -o -i ) > /tmp/squeeze-10-200_STDOUT_2618_16 2>&1" < /dev/null > /dev/null 2>&1
Bring remote stdout: scp pgxc@node4:/tmp/squeeze-10-200_STDOUT_2618_16 /tmp/STDOUT_2618_17 > /dev/null 2>&1
SET
SET
psql:/tmp/GENERAL_2618_15:12: ERROR:  role "pgxc" already exists
ALTER ROLE
REVOKE
REVOKE
GRANT
GRANT
CREATE NODE
CREATE NODE
CREATE NODE
CREATE NODE
CREATE NODE
CREATE NODE
You are now connected to database "postgres" as user "pgxc".
SET
SET
SET
SET
SET
COMMENT
CREATE EXTENSION
COMMENT
SET
SET
SET
psql:/tmp/GENERAL_2618_15:92: connection to server was lost
Actual Command: ssh pgxc@node4 "( pg_ctl stop -Z restoremode -D /opt/pgxc/nodes/coord ) > /tmp/squeeze-10-200_STDOUT_2618_18 2>&1" < /dev/null > /dev/null 2>&1
Bring remote stdout: scp pgxc@node4:/tmp/squeeze-10-200_STDOUT_2618_18 /tmp/STDOUT_2618_19 > /dev/null 2>&1
Starting coordinator master coord4
Done.
CREATE NODE
CREATE NODE
CREATE NODE
ALTER NODE

----------------------------------------------------------------

Error log
--------------------------------------------------------------------
ERROR:  role "pgxc" already exists
STATEMENT:  CREATE ROLE pgxc;
LOG:  server process (PID 3013) was terminated by signal 11: Segmentation fault
DETAIL:  Failed process was running: CREATE TABLE user_info_hash (
            id integer NOT NULL,
            firstname text,
            lastname text,
            info text
        )
        DISTRIBUTE BY HASH (id)
        TO NODE (datanode1,datanode2,datanode3);
LOG:  terminating any other active server processes
WARNING:  terminating connection because of crash of another server process
DETAIL:  The postmaster has commanded this server process to roll back the current transaction and exit, because another server process exited abnormally and possibly co
rrupted shared memory.
HINT:  In a moment you should be able to reconnect to the database and repeat your command.

---------------------------------------------------------------------------

Any comments are appreciated


Best Regards,

--  
Julian
使用 Sparrow (http://www.sparrowmailapp.com/?sig) 發信


On 2013年10月3日Thursday at 上午9:31, Koichi Suzuki wrote:

> You cannot add a coordinator in such a way.    There're many issued to be resolved internally.   You can configure and operate whole cluster with pgxc_ctl to get handy way to add coordinator/datanode.
>  
> I understand you have your cluster configured without pgxc_ctl.   In this case, adding coordinator manually could be a bit complicated work.   Sorry, I've not uploaded the detailed step to do it.  
>  
> Whole steps will be found in add_coordinatorMaster() function defined in coord_cmd.c of pgxc_ctl source code.   It will be found at contrib/pgxc_ctl in the release material.
>  
> Please allow a bit of time to find my time to upload this information to XC wiki.
>  
> Or, you can backup whole database with pg_dumpall, then reconfigure new xc cluster with additional coordinator, and then restore the backup.  
>  
> Regards;
>  
> ---
> Koichi Suzuki
>  
>  
>  
>  
> 2013/10/3 Julian <jul...@gm... (mailto:jul...@gm...)>
> > Dear Sir,
> >  
> > I have a cluster with 3 coordinators and 3 datanodes on 3 VM, today i
> > was try to added a new coordinator to the cluster, when i using command
> > "psql postgres -f coordinator.sql -p 5455" to restore the backup file to
> > the new coordinator.
> >  
> > Then i got this message :
> >  
> > psql:coordinator-dump.sql:105: connection to server was lost
> >  
> > In the log file:
> > ----------------------------------------------------------------------------------------------------------------------------------------------------------------
> > 2013-10-02 23:00:09.811 CST,,,29061,,524c34de.7185,2,,2013-10-02
> > 22:59:42 CST,,0,LOG,00000,"server process (PID 29094) was terminated by
> > signal 11: Segmentation fault","Failed process was r
> > unning: CREATE TABLE user_info_hash (
> >      id integer NOT NULL,
> >      firstname text,
> >      lastname text,
> >      info text
> > )
> > DISTRIBUTE BY HASH (id)
> > TO NODE (dn2,dn1,dn3);",,,,,,,,""
> > 2013-10-02 23:00:09.811 CST,,,29061,,524c34de.7185,3,,2013-10-02
> > 22:59:42 CST,,0,LOG,00000,"terminating any other active server
> > processes",,,,,,,,,""
> > ------------------------------------------------------------------------------------------------------------------------------------------------------------------
> > refer to
> > http://postgres-xc.sourceforge.net/docs/1_1/add-node-coordinator.html
> >  
> > Is there something i doing worng?
> >  
> >  
> > Thanks for your kindly reply.
> >  
> >   And sorry for my poor english.
> >  
> >  
> > Best regards.
> >  
> > ------------------------------------------------------------------------------
> > October Webinars: Code for Performance
> > Free Intel webinars can help you accelerate application performance.
> > Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from
> > the latest Intel processors and coprocessors. See abstracts and register >
> > http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk
> > _______________________________________________
> > Postgres-xc-general mailing list
> > Pos...@li... (mailto:Pos...@li...)
> > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general
>  

Sorry, I'm not familiar with Hibernate but JDBC does support DISTRIBUTE BY
REPLICATION.  You can issue CREATE TABLE through general jdbc apps, or you
can issue ALTER TABLE to change distributed table into replicated table
outside Hibernate, through psql.   There's no influence in other query
statement.

Regards;

---
Koichi Suzuki


2013/10/4 Anson Abraham <ans...@gm...>

> So I migrated a pg database (9.1) to postgres-xc 1.1.  Required me to
> essentially apply the Distribute By Replication to most of the tables.  But
> doing so, apparently this app threw out an error:
>
> Unable to upgrade schema to latest version.
> org.hibernate.exception.GenericJDBCException: ResultSet not positioned properly, perhaps you need to call next.
> 	at org.hibernate.exception.internal.StandardSQLExceptionConverter.convert(StandardSQLExceptionConverter.java:54)
> 	at org.hibernate.engine.jdbc.spi.SqlExceptionHelper.convert(SqlExceptionHelper.java:125)
> 	at org.hibernate.engine.jdbc.spi.SqlExceptionHelper.convert(SqlExceptionHelper.java:110)
> 	at org.hibernate.engine.jdbc.internal.proxy.AbstractResultSetProxyHandler.continueInvocation(AbstractResultSetProxyHandler.java:108)
> 	at org.hibernate.engine.jdbc.internal.proxy.AbstractProxyHandler.invoke(AbstractProxyHandler.java:81)
> 	at $Proxy10.getInt(Unknown Source)
> 	at com.cloudera.enterprise.dbutil.DbUtil.getSchemaVersion(DbUtil.java:212)
> 	at com.cloudera.enterprise.dbutil.DbUtil$1SchemaVersionWork.execute(DbUtil.java:159)
> 	at org.hibernate.jdbc.WorkExecutor.executeWork(WorkExecutor.java:54)
> 	at org.hibernate.internal.SessionImpl$2.accept(SessionImpl.java:1937)
> 	at org.hibernate.internal.SessionImpl$2.accept(SessionImpl.java:1934)
> 	at org.hibernate.engine.jdbc.internal.JdbcCoordinatorImpl.coordinateWork(JdbcCoordinatorImpl.java:211)
> 	at org.hibernate.internal.SessionImpl.doWork(SessionImpl.java:1955)
> 	at org.hibernate.internal.SessionImpl.doWork(SessionImpl.java:1941)
> 	at com.cloudera.enterprise.dbutil.DbUtil.getSchemaVersion(DbUtil.java:171)
> 	at com.cloudera.enterprise.dbutil.DbUtil.upgradeSchema(DbUtil.java:333)
> 	at com.cloudera.cmon.FhDatabaseManager.initialize(FhDatabaseManager.java:68)
> 	at com.cloudera.cmon.firehose.Main.main(Main.java:339)
> Caused by: org.postgresql.util.PSQLException: ResultSet not positioned properly, perhaps you need to call next.
> 	at org.postgresql.jdbc2.AbstractJdbc2ResultSet.checkResultSet(AbstractJdbc2ResultSet.java:2695)
> 	at org.postgresql.jdbc2.AbstractJdbc2ResultSet.getInt(AbstractJdbc2ResultSet.java:1992)
> 	at com.mchange.v2.c3p0.impl.NewProxyResultSet.getInt(NewProxyResultSet.java:2547)
> 	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> 	at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
> 	at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
> 	at java.lang.reflect.Method.invoke(Method.java:597)
> 	at org.hibernate.engine.jdbc.internal.proxy.AbstractResultSetProxyHandler.continueInvocation(AbstractResultSetProxyHandler.java:104)
> 	...
>
>
> So I'm assuming hibernate does not support Distribute by Replication?  Is that so, or did I not need to apply distribute by replication, though the table has a PK w/ is also reference as FK from another table?  If not, is there a hack to get around this, w/o having to recompile hibernate objects?
>
>
>
> ------------------------------------------------------------------------------
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most
> from
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk
> _______________________________________________
> Postgres-xc-general mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-general
>
>

On Thu, Oct 3, 2013 at 1:34 PM, Sandeep Gupta <gup...@gm...>wrote:

> Hi,
>
>
>  Setup has two computers. The configuration for gtm, gtm_proxy, datanode,
> and, coordinator all have listen_address='*'.
>
>
> Here are the info about setup:
> postgres=# select * from
> pgxc_node;                                                node_name    |
> node_type | node_port | node_host | nodeis_primary | nodeis_preferred |
> node_id
>
> ----------------+-----------+-----------+-----------+----------------+------------------+-------------
>  coord1         | C         |      5432 | localhost | f              |
> f                |  1885696643
>  datanode_c1_d1 | D         |     45421 | sfx057    | f              |
> f                | -1199687708
>  datanode_c2_d1 | D         |     45421 | sfx050    | f              |
> f                |  -294121722
> (3 rows)
>
>
> select pgxc_pool_reload();
>  pgxc_pool_reload
> ------------------
>  t
> (1 row)
>
> However, when I create a table I get
>
> ERROR:  Failed to get pooled connections
>

Please check if you have a firewall running and open up the necessary ports.

Also, edit pg_hba.conf to allow connections from the coordinator to the
data nodes.


>
> Is there anything else apart from listen_address that I am missing?
> Please let me know.
>
>
> -Sandeep
>
>
> ------------------------------------------------------------------------------
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most
> from
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60134791&iu=/4140/ostg.clktrk
> _______________________________________________
> Postgres-xc-general mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-general
>
>


-- 
Mason Sharp

StormDB - http://www.stormdb.com
The Database Cloud
Postgres-XC Support and Services