Hi,
As far as I can tell, manual intervention on a log-shipping node is
required if one adds a new set to a cluster.
for a new set with:
tab_id | tab_reloid | tab_relname | tab_nspname | tab_set |
tab_idxname | tab_altered | tab_comment
--------+------------+-------------------+-------------+---------+------------------------+-------------+--------------------------------
106 | 44169 | new_table | public | 5 |
new_table_pkey | t | table public.new_table
(1 row)
seq_id | seq_reloid | seq_relname | seq_nspname | seq_set |
seq_comment
--------+------------+-----------------------+-------------+---------+---------------------------------------
127 | 44167 | new_table_seq | public | 5 | Sequence
public.new_table_seq
(1 row)
Pairing down slony1_dump.sh to fetch only the new set, I get
copy "_oxrsorg".sl_sequence_offline from stdin;
127 new_table_seq public
\.
select "_cluster".sequenceSetValue_offline(127, '1');
insert into "_oxrsorg".sl_setsync_offline values (5, '41336');
copy public.test_slonik_sleep
("id","dummy_field1","dummy_field2","dummy_field3") from stdin;
\.
commit;
Does this seem reasonable?
Thx
JP
--
JP Fletcher
Database Administrator
Afilias Canada
voice: 416.646.3304 ext. 123
fax: 416.646.3305
mobile: 416.561.4763
[EMAIL PROTECTED]
_______________________________________________
Slony1-general mailing list
[email protected]
http://gborg.postgresql.org/mailman/listinfo/slony1-general
