[Pgcluster-general] read_packet():unexpected EOF

Marcin Gabrowski marcin at gabrowski.pl
Fri Apr 25 17:02:03 UTC 2008 

I'm sure this cluster is alive. I can by psql login into psql.
On recovering cluster db in logs is ok.
1st recovery step of [base] directory...OK
1st recovery step of [pg_clog] directory...OK
1st recovery step of [pg_xlog] directory...OK
1st sync_table_space OK
2nd recovery step of [global] directory...OK
2nd recovery step of [base] directory...OK
2nd recovery step of [pg_clog] directory...OK
2nd recovery step of [pg_xlog] directory...OK
2nd sync_table_space OK
2nd recovery successed

But on master for recovering I get:  read_packet():unexpected EOF

On 2008-04-23, at 11:23, <mitani at sraw.co.jp> <mitani at sraw.co.jp> wrote:

> Hi,
>
> It seems cluster DB is died.
> Could you chekc log of cluster DB and OS.
>
> Regards,
> --------------
> At.Mitnai
>
> -- original message --
> From: Marcin Gabrowski<marcin at gabrowski.pl>
> To: <pgcluster-general at pgfoundry.org>
> Sent: Tue, 22 Apr 2008 18:04:17 +0200
> Subject: [Pgcluster-general] read_packet():unexpected EOF
>
>> Hi,
>>
>> During recovery from 1st node I got " read_packet():unexpected  
>> EOF". I
>> think that is bug.
>>
>> My config is as follows (http://article.gmane.org/gmane.comp.db.postgresql.pgcluster/1065
>> )
>> We're testing togather pgcluster. Full debug I attached below:
>>
>> Tue Apr 22 18:00:02 2008   
>> PGRsend_replicate_packet_to_server():connect
>> db:template1 port:5830 user:pgcluster host:172.30.200.21 query:VACUUM
>> Tue Apr 22 18:00:02 2008
>> send_replicate_packet_to_server():sync_command(SELECT
>> PGR_SYSTEM_COMMAND_FUNCTION(3,0,0,0,1,247) )
>> Tue Apr 22 18:00:03 2008  send_replicate_packet_to_server():PQexec
>> send :VACUUM
>> Tue Apr 22 18:00:03 2008  send_replicate_packet_to_server():PQexec
>> returns :VACUUM
>> Tue Apr 22 18:00:03 2008  first_setup_recovery():send packet to  
>> master
>> cluster_1 recoveryPort 7001
>> Tue Apr 22 18:00:03 2008  first_setup_recovery():wait answer from
>> master server
>> Tue Apr 22 18:00:03 2008  read_packet():receive packet
>> Tue Apr 22 18:00:03 2008  no = 3
>> Tue Apr 22 18:00:03 2008  max_connect = 103
>> Tue Apr 22 18:00:03 2008  port = 5830
>> Tue Apr 22 18:00:03 2008  recoveryPort = 7001
>> Tue Apr 22 18:00:03 2008  hostName = db-01
>> Tue Apr 22 18:00:03 2008  pg_data = /home/pgcluster/data
>> Tue Apr 22 18:00:03 2008  first_setup_recovery():get answer from
>> master:no[3]
>> Tue Apr 22 18:00:03 2008  PGRsend_load_balance_packet():host[lb_3]
>> port[6001]
>> Tue Apr 22 18:00:03 2008  pgrecovery_loop():first_setup_recovery  
>> end:0
>> Tue Apr 22 18:00:03 2008  PGRdo_replicate():query ::
>> Tue Apr 22 18:00:03 2008  cmdSts=P
>> Tue Apr 22 18:00:03 2008  cmdType=P
>> Tue Apr 22 18:00:03 2008  rlog=0
>> Tue Apr 22 18:00:03 2008  port=5830
>> Tue Apr 22 18:00:03 2008  pid=23153
>> Tue Apr 22 18:00:03 2008  from_host=172.30.200.21
>> Tue Apr 22 18:00:03 2008  dbName=bossa
>> Tue Apr 22 18:00:03 2008  userName=eogroup
>> Tue Apr 22 18:00:03 2008  recieve sec=1208880003
>> Tue Apr 22 18:00:03 2008  recieve usec=871647
>> Tue Apr 22 18:00:03 2008  query_size=57
>> Tue Apr 22 18:00:03 2008  request_id=3153
>> Tue Apr 22 18:00:03 2008  replicate_id=0
>> Tue Apr 22 18:00:03 2008  recovery_status=1
>> Tue Apr 22 18:00:03 2008  sem_lock [1] req
>> Tue Apr 22 18:00:03 2008  sem_lock [1] got it
>> Tue Apr 22 18:00:03 2008  source host
>> Tue Apr 22 18:00:03 2008  end thread_send_source()
>> Tue Apr 22 18:00:03 2008  sem_unlock[1]
>> Tue Apr 22 18:00:03 2008  PGRdo_replicate():query ::
>> Tue Apr 22 18:00:03 2008  cmdSts=P
>> Tue Apr 22 18:00:03 2008  cmdType=B
>> Tue Apr 22 18:00:03 2008  rlog=0
>> Tue Apr 22 18:00:03 2008  port=5830
>> Tue Apr 22 18:00:03 2008  pid=23153
>> Tue Apr 22 18:00:03 2008  from_host=172.30.200.21
>> Tue Apr 22 18:00:03 2008  dbName=bossa
>> Tue Apr 22 18:00:03 2008  userName=eogroup
>> Tue Apr 22 18:00:03 2008  recieve sec=1208880003
>> Tue Apr 22 18:00:03 2008  recieve usec=873123
>> Tue Apr 22 18:00:03 2008  query_size=9
>> Tue Apr 22 18:00:03 2008  request_id=3154
>> Tue Apr 22 18:00:03 2008  replicate_id=0
>> Tue Apr 22 18:00:03 2008  recovery_status=3
>> Tue Apr 22 18:00:03 2008  sem_lock [1] req
>> Tue Apr 22 18:00:03 2008  sem_lock [1] got it
>> Tue Apr 22 18:00:03 2008  replicate_packet_send_internal() master is
>> using for recovery
>> Tue Apr 22 18:00:03 2008  sem_unlock[1]
>> Tue Apr 22 18:00:07 2008  read_packet():receive packet
>> Tue Apr 22 18:00:07 2008  no = 5
>> Tue Apr 22 18:00:07 2008  max_connect = 103
>> Tue Apr 22 18:00:07 2008  port = 5830
>> Tue Apr 22 18:00:07 2008  recoveryPort = 7001
>> Tue Apr 22 18:00:07 2008  hostName = cluster_1
>> Tue Apr 22 18:00:07 2008  pg_data = /home/pgcluster/data
>> Tue Apr 22 18:00:07 2008  pgrecovery_loop():receive packet no:5
>> Tue Apr 22 18:00:07 2008  pgrecovery_loop():2nd master cluster_1 -  
>> 5830
>> Tue Apr 22 18:00:07 2008  pgrecovery_loop():2nd target db-02 - 5830
>> Tue Apr 22 18:00:08 2008   
>> PGRsend_replicate_packet_to_server():connect
>> db:template1 port:5830 user:pgcluster host:172.30.200.21 query:VACUUM
>> FULL
>> Tue Apr 22 18:00:08 2008
>> send_replicate_packet_to_server():sync_command(SELECT
>> PGR_SYSTEM_COMMAND_FUNCTION(3,0,0,0,1,249) )
>> Tue Apr 22 18:00:09 2008  send_replicate_packet_to_server():PQexec
>> send :VACUUM FULL
>> Tue Apr 22 18:00:09 2008  send_replicate_packet_to_server():PQexec
>> returns :VACUUM
>> Tue Apr 22 18:00:09 2008  read_packet():receive packet
>> Tue Apr 22 18:00:09 2008  no = 7
>> Tue Apr 22 18:00:09 2008  max_connect = 103
>> Tue Apr 22 18:00:09 2008  port = 5830
>> Tue Apr 22 18:00:09 2008  recoveryPort = 7001
>> Tue Apr 22 18:00:09 2008  hostName = db-01
>> Tue Apr 22 18:00:09 2008  pg_data = /home/pgcluster/data
>> Tue Apr 22 18:00:09 2008  pgrecovery_loop():second_setup_recovery  
>> end :0
>> Tue Apr 22 18:00:10 2008  read_packet():unexpected EOF
>> Tue Apr 22 18:00:10 2008  pgrecovery_loop():receive packet no:8
>>
>>
>> -- 
>> gg: 2600 ripe: mg3051 jid: gaber/gentoo.pl http://gabrowski.pl/
>>
>>
>> _______________________________________________
>> Pgcluster-general mailing list
>> Pgcluster-general at pgfoundry.org
>> http://pgfoundry.org/mailman/listinfo/pgcluster-general
>>
>
> _______________________________________________
> Pgcluster-general mailing list
> Pgcluster-general at pgfoundry.org
> http://pgfoundry.org/mailman/listinfo/pgcluster-general

-- 
gg: 2600 ripe: mg3051 jid: gaber/gentoo.pl http://gabrowski.pl/

More information about the Pgcluster-general mailing list