Hello, The attached patch tries to improve the docs for the implications of what "SYNC aborted" means. A peer review for accuracy would be appreciated.
At one one I was given direct CVS access to make doc commits, but I couldn't find my access information for that today. Mark
RCS file: /slony1/slony1-engine/doc/adminguide/loganalysis.sgml,v retrieving revision 1.8 diff -r1.8 loganalysis.sgml 697c697,700 < <para> If any errors have been encountered that haven't already aborted the <command>SYNC</command>, this catches and aborts it.</para></listitem> --- > <para>This aborts the <command>SYNC</command>. The same sync may be tried again in the near future. If it continues to fail, > replication may never catch up without intervention. It may be necessary to unsubscribe/resubscribe the affected slave set, > or if there is only one set on the slave node, drop and recreate the slave node, which will have better performance. If application connections can be moved to the master during this time, downtime may not be necessary. > </para></listitem>
_______________________________________________ Slony1-general mailing list [email protected] http://gborg.postgresql.org/mailman/listinfo/slony1-general
