Hi,
After viewing the pgreplicate.log, it seem the replication server will
update or delete records using the OID, isn't it?
If that's the case, it may be problem to sync. records in each node because
OID always be different during backup/restoration or import data.
Mike
----- Original Message -----
From: "mike" <mike.ho@xxxxxxxxxxxxxxx>
To: <pgcluster-general@xxxxxxxxxxxxx>
Sent: Tuesday, March 06, 2007 10:45 AM
Subject: Re: [Pgcluster-general] Problem in Pgcluster
> Hi,
>
> I'm not sure if it's a bug or not:
>
> load balancer: port 5431 ip : 128.128.3.1
> 1st node : port 5432 ip : 128.128.3.1
> 2nd node : port 5432 ip : 128.128.3.2
>
> When i delete some records from table in 5431 port, only 1st node take
> effect and records remains in 2nd node.
>
> From: "mike" <mike.ho@xxxxxxxxxxxxxxx>
> To: <pgcluster-general@xxxxxxxxxxxxx>
> Sent: Tuesday, March 06, 2007 10:27 AM
> Subject: Re: [Pgcluster-general] Problem in Pgcluster
>
>
> > Hi,
> >
> > I have 2 nodes installed. First node have load balancer, replication
> server
> > and postgres db install. Seconde node only have postgres db installed.
> >
> > Here's the log in pgreplicate.log after i shutdown the first postgres db
:
> >
> > Tue Mar 6 10:19:22 2007 recieve sec=1173147562
> > Tue Mar 6 10:19:22 2007 recieve usec=910395
> > Tue Mar 6 10:19:22 2007 query_size=21
> > Tue Mar 6 10:19:22 2007 request_id=1
> > Tue Mar 6 10:19:22 2007 replicate_id=0
> > Tue Mar 6 10:19:22 2007 recovery_status=0
> > Tue Mar 6 10:19:22 2007 query=PGR_CLOSE_CONNECTION
> > Tue Mar 6 10:19:22 2007 sem_lock [1] req
> > Tue Mar 6 10:19:22 2007 sem_lock [1] got it
> > Tue Mar 6 10:19:22 2007 replicate_packet_send_internal():connect
> db:cinema
> > por
> > t:5432 user:cinema host:128.128.3.2 query:PGR_CLOSE_CONNECTION
> > Tue Mar 6 10:19:22 2007 start thread_send_cluster()
> > Tue Mar 6 10:19:22 2007 sem_unlock[1]
> >
> > and pglb.log
> >
> > Tue Mar 6 10:19:10 2007 PGRscan_cluster:0 ClusterDB can be used
> > Tue Mar 6 10:19:10 2007 PGRscan_cluster:128.128.3.1 [5432],useFlag->99
> > max->32
> > use_num->1
> >
> > Tue Mar 6 10:19:10 2007 PGRscan_cluster:0 ClusterDB can be used
> > Tue Mar 6 10:19:10 2007 PGRscan_cluster:128.128.3.1 [5432],useFlag->99
> > max->32
> > use_num->1
> >
> > Tue Mar 6 10:19:10 2007 PGRload_balance():no cluster available
> > Tue Mar 6 10:19:10 2007 load_balance_main():load balance process
failed
> > Tue Mar 6 10:19:10 2007 load_balance_main():no cluster available
> > Tue Mar 6 10:19:10 2007 do_accept():I am 16191 accept fd 6
> > Tue Mar 6 10:19:10 2007 read_startup_packet():Protocol Major: 2 Minor:
0
> > datab
> > ase: cinema user: cinema
> >
> > It seem that the db cluster will not available after the postgres db in
> > first node is down;
> > what's the reason?
> >
> > (Suppose clusting db should still be available in this case).
> >
> >
> > ----- Original Message -----
> > From: <a.mitani@xxxxxxxxxxxxxx>
> > To: <pgcluster-general@xxxxxxxxxxxxx>
> > Sent: Saturday, March 03, 2007 5:00 PM
> > Subject: Re: [Pgcluster-general] Problem in Pgcluster
> >
> >
> > > Hi,
> > >
> > > You can save error messages into log file start server with "-l"
option.
> > > And you can save debug message into log file start server with "-v"
> > option.
> > >
> > > You can find all option when start server with "-h" option.
> > >
> > > Regards,
> > > -------------------
> > > At.Mitani
> > >
> > > > Hi At.Mitani,
> > > >
> > > > No debug message found in any log files.
> > > > Now i extract the tar file for 1.7rc4, how can i apply the patch ?
> > > >
> > > > ----- Original Message -----
> > > > From: "mike" <mike.ho@xxxxxxxxxxxxxxx>
> > > > To: <pgcluster-general@xxxxxxxxxxxxx>
> > > > Sent: Saturday, March 03, 2007 9:32 AM
> > > > Subject: Re: [Pgcluster-general] Problem in Pgcluster
> > > >
> > > >
> > > >> Hi At.Mitani,
> > > >>
> > > >> Where can i view the debug message?
> > > >>
> > > >> Now i set Use_Replication_Log to 'No'
> > > >>
> > > >>
> > > >> Mike
> > > >>
> > > >>
> > > >> ----- Original Message -----
> > > >> From: <a.mitani@xxxxxxxxxxxxxx>
> > > >> To: <pgcluster-general@xxxxxxxxxxxxx>
> > > >> Sent: Friday, March 02, 2007 8:24 PM
> > > >> Subject: Re: [Pgcluster-general] Problem in Pgcluster
> > > >>
> > > >>
> > > >> > Hi Mike,
> > > >> >
> > > >> > One more, please set "no" to "<Use_Replication_Log>" in your
> > > >> > pgreplicate.conf.
> > > >> >
> > > >> > And please show us your debug message of pglb and pgreplication
as
> > > >> well,
> > > >> > if you can.
> > > >> >
> > > >> > Regards,
> > > >> > ---------------------
> > > >> > At.Mitani
> > > >> >
> > > >> >
> > > >> > > Thank for your reply. Both server can listen to all to all ip
in
> > > >> > > pg_hba.conf
> > > >> > > (All trust)
> > > >> > >
> > > >> > > listen_addresses = '*' in postgres config. file.
> > > >> > >
> > > >> > > Mike
> > > >> > >
> > > >> > > ----- Original Message -----
> > > >> > > From: <a.mitani@xxxxxxxxxxxxxx>
> > > >> > > To: <pgcluster-general@xxxxxxxxxxxxx>
> > > >> > > Sent: Friday, March 02, 2007 7:18 PM
> > > >> > > Subject: Re: [Pgcluster-general] Problem in Pgcluster
> > > >> > >
> > > >> > >
> > > >> > >> Hi,
> > > >> > >>
> > > >> > >> At first, please discribe one data on one line in all
> > configuration
> > > >> > >> files.
> > > >> > >> Next, please check "listen_addresses" in "postgresql.conf" and
> > > >> > >> "pg_hba.conf" on each Cluster DB.
> > > >> > >>
> > > >> > >> Regards,
> > > >> > >> ----------------
> > > >> > >> At.Mitani
> > > >> > >>
> > > >> > >>
> > > >> > >> > Dear all, I'm trying to use pgcluster 1.7.0-rc4 (with
> postgresql
> > > >> > > 8.2.3)My
> > > >> > >> > setup: - two postgresql servers, 128.128.3.1 and .2
> > > >> > >> > Load balancer an replication ride on 128.128.3.1 serverThe
> > > >> > >> > pgreplicate.conf file looks like this:
> > > >> > >> >
> > > >> > >> > ## pgreplicate.conf begin
> > > >> > >> > <Cluster_Server_Info> <Host_Name> 128.128.3.2
> </Host_Name>
> > > >> > > <Port>
> > > >> > >> > 5432 </Port> <Recovery_Port>
> 7001
> > > >> > >> > </Recovery_Port></Cluster_Server_Info><Cluster_Server_Info>
> > > >> > > <Host_Name>
> > > >> > >> > 128.128.3.1 </Host_Name> <Port> 5432
> > > >> > >> </Port>
> > > >> > >> > <Recovery_Port> 7001
> > > >> > >> >
> </Recovery_Port></Cluster_Server_Info><LoadBalance_Server_Info>
> > > >> > >> > <Host_Name> 128.128.3.1 </Host_Name>
<Recovery_Port>
> > > >> > > 6001
> > > >> > >> >
> </Recovery_Port></LoadBalance_Server_Info><Host_Name>
> > > >> > >> > 128.128.3.1 </Host_Name><Replication_Port> 8001
> > > >> > >> > </Replication_Port><Recovery_Port> 8101
> > > >> > >> > </Recovery_Port><RLOG_Port> 8301
> > > >> > >> > </RLOG_Port><Response_Mode> normal
> > > >> > >> > </Response_Mode><Use_Replication_Log> yes
> > > >> > >> > </Use_Replication_Log><Replication_Timeout> 1min
> > > >> > >> > </Replication_Timeout><LifeCheck_Timeout> 3s
> > > >> > >> > </LifeCheck_Timeout><LifeCheck_Interval> 15s
> > > >> > >> > </LifeCheck_Interval>
> > > >> > >> > ## === pgreplicate.conf end
> > > >> > >> > The cluster.conf file for the two servers is identical:
> > > >> > >> > ## === cluster.conf file
> > > >> > >> > <Replicate_Server_Info>
> > > >> > >> > <Host_Name> 128.128.3.1 </Host_Name>
> > > >> > >> > <Port> 8001 </Port>
> > > >> > >> > <Recovery_Port> 8101 </Recovery_Port>
> > > >> > >> > </Replicate_Server_Info>
> > > >> > >> > <Host_Name> 128.128.3.1 </Host_Name>
> > > >> > >> > <Recovery_Port> 7001
> </Recovery_Port>
> > > >> > >> > <Rsync_Path> /usr/bin/rsync
</Rsync_Path>
> > > >> > >> > <Rsync_Option> ssh -1
> </Rsync_Option>
> > > >> > >> > <Rsync_Compress> yes
> > </Rsync_Compress>
> > > >> > >> > <Pg_Dump_Path> /usr/local/pgsql/bin/pg_dump
> > > >> > >> > </Pg_Dump_Path>
> > > >> > >> > <When_Stand_Alone> read_only
> > > >> </When_Stand_Alone>
> > > >> > >> > <Replication_Timeout> 1min
</Replication_Timeout>
> > > >> > >> > <LifeCheck_Timeout> 3s </LifeCheck_Timeout>
> > > >> > >> > <LifeCheck_Interval> 11s
</LifeCheck_Interval>
> > > >> > >> >
> > > >> > >> > ## === end of cluster.conf file
> > > >> > >> >
> > > >> > >> > And load balancer config:
> > > >> > >> > <Cluster_Server_Info>
> > > >> > >> > <Host_Name> 128.128.3.1 </Host_Name>
> > > >> > >> > <Port> 5432 </Port>
> > > >> > >> > <Max_Connect> 32 </Max_Connect>
> > > >> > >> > </Cluster_Server_Info>
> > > >> > >> > <Cluster_Server_Info>
> > > >> > >> > <Host_Name> 128.128.3.2 </Host_Name>
> > > >> > >> > <Port> 5432 </Port>
> > > >> > >> > <Max_Connect> 32 </Max_Connect>
> > > >> > >> > </Cluster_Server_Info>
> > > >> > >> >
> > > >> > >> > <Host_Name> 128.128.3.1 </Host_Name>
> > > >> > >> > <Backend_Socket_Dir> /tmp </Backend_Socket_Dir>
> > > >> > >> > <Receive_Port> 5431 </Receive_Port>
> > > >> > >> > <Recovery_Port> 6001 </Recovery_Port>
> > > >> > >> > <Max_Cluster_Num> 128 </Max_Cluster_Num>
> > > >> > >> > <Use_Connection_Pooling> no </Use_Connection_Pooling>
> > > >> > >> > <LifeCheck_Timeout> 3s
> > > >> > >> > </LifeCheck_Timeout>
> > > >> > >> > <LifeCheck_Interval> 15s
> > > > </LifeCheck_Interval>
> > > >> > >> >
> > > >> > >> >
> > > >> > >> > Load balanacer list 5431 port and sync. the structure and
data
> > to
> > > >> > >> > 128.128.3.1 & .2 database
> > > >> > >> >
> > > >> > >> > I want to test on the situation when 1 of cluster server
down.
> > > >> > >> > However, after i shutdown postgres in 3.1 server, the whole
> > > > postgres
> > > >> > >> was
> > > >> > >> > unable to access
> > > >> > >> > with error 'Sorry, backend connection is full'
> > > >> > >> >
> > > >> > >> > Alternatively, if i shutdown postgres in 3.2, load balancer
> can
> > > > still
> > > >> > >> > listen the request and propagate result to database in 3.1
> > > >> > >> > server; But when the 3.2 postgres resume, the latest change
> > > >> cannot
> > > > be
> > > >> > >> > replicated to 3.2 database.
> > > >> > >> >
> > > >> > >> > Any clue?
> > > >> > >> >
> > > >> > >> > Mike
> > > >> > >> >
> > > >> > >> >
> > > >> > >> >
> > > >> > >> >
> > > >> > >> >
> > > >> > >> >
> > > >> > >> >
> > > >> > >> >
> > > >> > >> > _______________________________________________
> > > >> > >> > Pgcluster-general mailing list
> > > >> > >> > Pgcluster-general@xxxxxxxxxxxxx
> > > >> > >> > http://pgfoundry.org/mailman/listinfo/pgcluster-general
> > > >> > >> >
> > > >> > >>
> > > >> > >> _______________________________________________
> > > >> > >> Pgcluster-general mailing list
> > > >> > >> Pgcluster-general@xxxxxxxxxxxxx
> > > >> > >> http://pgfoundry.org/mailman/listinfo/pgcluster-general
> > > >> > >>
> > > >> > >
> > > >> > >
> > > >> > > _______________________________________________
> > > >> > > Pgcluster-general mailing list
> > > >> > > Pgcluster-general@xxxxxxxxxxxxx
> > > >> > > http://pgfoundry.org/mailman/listinfo/pgcluster-general
> > > >> > >
> > > >> >
> > > >> > _______________________________________________
> > > >> > Pgcluster-general mailing list
> > > >> > Pgcluster-general@xxxxxxxxxxxxx
> > > >> > http://pgfoundry.org/mailman/listinfo/pgcluster-general
> > > >> >
> > > >>
> > > >
> > > >
> > > > _______________________________________________
> > > > Pgcluster-general mailing list
> > > > Pgcluster-general@xxxxxxxxxxxxx
> > > > http://pgfoundry.org/mailman/listinfo/pgcluster-general
> > > >
> > >
> > > _______________________________________________
> > > Pgcluster-general mailing list
> > > Pgcluster-general@xxxxxxxxxxxxx
> > > http://pgfoundry.org/mailman/listinfo/pgcluster-general
> > >
> >
>
|
