I am running a slony cluster against three built-from-src postgres-8.1.4 =

servers. I had been running 1.2.6 successfully.

After installing the new slony, I then restarted each of the databases =

to make sure they picked up the new libs.

But after issuing the updgrade commands:

    cluster name =3D $CLUSTERNAME;
    node 1 admin conninfo =3D 'dbname=3D$SRV1DBNAME port=3D$SRV1PORT =

host=3D$SRV1HOST user=3D$REPLICATIONUSER';
    node 2 admin conninfo =3D 'dbname=3D$SRV2DBNAME port=3D$SRV2PORT =

host=3D$SRV2HOST user=3D$REPLICATIONUSER';
    node 3 admin conninfo =3D 'dbname=3D$SRV3DBNAME port=3D$SRV3PORT =

host=3D$SRV3HOST user=3D$REPLICATIONUSER';

    update functions (id =3D 1);
    update functions (id =3D 2);
    update functions (id =3D 3);

    restart node 1;
    restart node 2;
    restart node 3;

Then restarting a slon daemon on node 3, I see the following fatal error:

May 12 00:41:47 srv3 slon[8573]: [42-1] 2007-05-12 00:41:47 CDT CONFIG =

main: configuration complete - starting threads
May 12 00:41:47 srv3 postgres[8587]: [64-1] NOTICE:  Slony-I: cleanup =

stale sl_nodelock entry for pid=3D8440
May 12 00:41:47 srv3 slon[8573]: [44-1] 2007-05-12 00:41:47 CDT CONFIG =

enableNode: no_id=3D1
May 12 00:41:48 srv3 slon[8573]: [47-1] 2007-05-12 00:41:47 CDT CONFIG =

enableNode: no_id=3D2
May 12 00:41:50 srv3 slon[8573]: [54-1] 2007-05-12 00:41:50 CDT FATAL  =

syncThread: "select =

"_srv1_srv2_cluster".createEvent('_srv1_srv2_cluster', 'SYNC', NULL);" -
May 12 00:41:50 srv3 slon[8573]: [54-2]  server closed the connection =

unexpectedly
May 12 00:41:50 srv3 slon[8573]: [54-3]         This probably means the =

server terminated abnormally
May 12 00:41:50 srv3 slon[8573]: [54-4]         before or while =

processing the request.
May 12 00:41:50 srv3 slon[8573]: [58-1] 2007-05-12 00:41:50 CDT INFO   =

remoteListenThread_1: disconnecting from 'dbname=3Daltos_research port=3D54=
33
May 12 00:41:50 srv3 slon[8573]: [58-2]  host=3Dsrv1.altosresearch.com =

user=3Dpostgres'
May 12 00:41:50 srv3 postgres[7320]: [64-1] LOG:  server process (PID =

8597) was terminated by signal 11
May 12 00:41:50 srv3 postgres[7320]: [65-1] LOG:  terminating any other =

active server processes

What is the most likely cause of this error?

Thanks,
Jason

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.slony.info/pipermail/slony1-general/attachments/20070511/=
5d475a7d/attachment.htm