Thu Jun 12 11:26:34 PDT 2008
- Previous message: [Slony1-general] Add new slave to cluster
- Next message: [Slony1-general] order of replication
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Hi, While trying to add tables to a replication, I had a strange behaviour I can't repair by now. I use slony-1.2.13, PG. 8.2.7. I have 2 sets replicating in both ways. One replication goes from 1 -> 11 -> 71 -> 72 the other one from 72 to 1. What I did was : create set id 99 originating on 1 add tables to that set subscribe 11 to 99 from 1, 71 to 99 from 11 and 72 to 99 from 71. But then I had the message that set 99 did not exist on 72, and replication hanged. Trying to drop set 99 from replication did not let me resync node 72, as it still tried to subscribe to unexisting set 99. I then switch set 2 from 72 to 71. That worked. Then I tried to uninstall node 72, drop schema. OK But from then, I couldn't issue dropnode( 72 ) on node 1. It seems to be waiting indefinitely, and I have such messages in logs : Jun 12 20:22:54 myhost slon[29751]: [138-1] 2008-06-12 20:22:54 CEST ERROR remoteListenThread_72: "select "_replication".registerNodeConnection(1); listen "_replication_Event"; Jun 12 20:22:54 myhost slon[29751]: [138-2] " - ERREUR: le schéma « _replication » n'existe pas (it complains about replication schema not existing)... What can I do as I do not want to recreate the whole replication from scratch (many nodes, two ways, two sets...). What's more recrreating a single node does take almost 3 hours. Thanks by advance. Regards -- Stéphane SCHILDKNECHT Président de PostgreSQLFr Tél. 09 53 69 97 12 http://www.postgresql.fr
- Previous message: [Slony1-general] Add new slave to cluster
- Next message: [Slony1-general] order of replication
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-general mailing list