Hi,
Tnx for your help.
 
The scenario is very simple : after I set upt the cluster (which works just
fine) , i'm trying three different things , and got three different errors:
1. execute the "failover" script  on the slave , just like documented : i
got the specific error (...INSERT has more expressions than target
columns...)
2. execute the "controlled switchover" on the slave , i got an error with
the event ...
3. i'm trying to add an entry to one of my tables (which has
primarykey,foreignkey and  "serial" as ID) - i get the error that the
primary key was violated...
 
I'm pretty lost with it, i tried the same with Slony2.0.0 (stabe) and with
latest postgers (8.3.5) and got the same behaviour.I'm probably missing
something...
 
One more word: when i'm setting up the cluster i'm taking the following
procedure:
* pg_dumpall from master to slave
* creating the slave on the master (slonik scripts , subscribing)
* start slon on slave
* start slon on master
 
I believe that it should be fine , becase the cluster is working just
fine...
 
anyhow, tnx for your help.
tnx,
zvika

Rafael Domiciano wrote:
> 
> You didn't specified in what server you removed the cluster.Maybe the
> sequences on your remote server are all outdated.
> 
> 2008/12/4 zferentz <zferentz at gmail.com>
> 
>>
>> Hi,
>>
>> (First, i'd like to apologize , i've accidently submited this question to
>> another group...)
>>
>> I've encountered two different problems that related to simple
>> master/slave
>> topology, i'd appreciate any help...
>>
>> First problem:
>> my replication works just fine  (single master , single slave on remote
>> machine) . however - when  I remove the cluster (schema) either by GUI
>> (pgAdmin ) or by "DROP" script or by slonik script, i cannot add new
>> items
>> to my tables. i get "duplicate key value violates unique constraint" with
>> my
>> primary-key.
>> I've noticed that the problem starts the moment i start the "slon"
>> script,
>> and the problem doesn't disappear even when i stopped/remove the slony
>> cluster.
>>
>> Second problem:
>> when i'm executing the "failover" script (as described in the
>> documetnation)
>> i get some  weird errors :
>>
>>     ERROR:  INSERT has more expressions than target columns
>>      CONTEXT:  SQL statement "INSERT INTO "_esa_cluster".sl_event
>> (ev_origin,         ev_seqno, ev_timestamp, ev_snapshot, ev_type,
>> ev_data1,
>> ev_data2, ev_data3, ev_data4)      values ( $1 ,
>> "pg_catalog".nextval('"_esa_cluster".sl_event_seq'), CURRENT_TIMESTAMP,
>> '0',
>> '0', '0:0:', 'ACCEPT_SET',  $2 ::text,  $3 ::text,  $1 ::text,  $4
>> ::text)"
>>     PL/pgSQL function "failoverset_int" line 35 at SQL statement
>>     SQL statement "SELECT  "_esa_cluster".failoverSet_int( $1 ,  $2 ,  $3
>> ,
>> $4 )"
>>     PL/pgSQL function "failednode2" line 39 at PERFORM
>>
>> The net result of the two problems is that the slave is not usable , so i
>> cannot use slony...
>>
>> I'm using Debian ,and everything was built from sourcecode
>> (postgres-8.3.4
>> ,
>> slony1-2.0.0-rc2)
>>
>>
>> Any help would be appreciated...
>> --
>> View this message in context:
>> http://www.nabble.com/Slony-%282.0.0rc2%29-SLAVE-Failures-...-Help....-tp20829231p20829231.html
>> Sent from the Slony-I -- General mailing list archive at Nabble.com.
>>
>> _______________________________________________
>> Slony1-general mailing list
>> Slony1-general at lists.slony.info
>> http://lists.slony.info/mailman/listinfo/slony1-general
>>
> 
> _______________________________________________
> Slony1-general mailing list
> Slony1-general at lists.slony.info
> http://lists.slony.info/mailman/listinfo/slony1-general
> 
> 

-- 
View this message in context: http://www.nabble.com/Slony-%282.0.0rc2%29-SLAVE-Failures-...-Help....-tp20829231p20889792.html
Sent from the Slony-I -- General mailing list archive at Nabble.com.