<br />I noticed that pg_attribute has rows with the same OID! You can <br />verify this by running. <p># select
count(oid),oidfrom pg_attribute group by oid having <br />count(oid) > 1; <p>We see that these duplicate OIDs appear
tohappen when <br />pg_attibute rows are inserted for indexes and primary keys. <p>select relname, attname <br />from
pg_attribute, pg_class <br />where attrelid = pg_class.oid and <br />pg_attribute.oid in (select oid <br
/> from pg_attribute <br /> group by oid <br /> having count(oid) > 1);
<p>Isee: <br /><tt> relname | attname</tt><br /><tt>
---------------------------------+--------------</tt><br/><tt> pg_aggregate_name_type_index |
aggname</tt><br/><tt> pg_aggregate_name_type_index | aggbasetype</tt><br /><tt>
pg_amop_opid_index | amopclaid</tt><br /><tt> pg_amop_opid_index | amopopr</tt><br
/><tt> pg_amop_opid_index | amopid</tt><br /><tt> pg_amop_strategy_index |
amopid</tt><br/><tt> pg_amop_strategy_index | amopclaid</tt><br /><tt>
pg_amop_strategy_index | amopstrategy</tt><br /><tt> pg_attribute_relid_attnam_index | attrelid</tt><br
/><tt> pg_attribute_relid_attnam_index | attname</tt><br /><tt> pg_attribute_relid_attnum_index |
attrelid</tt><br/><tt> pg_attribute_relid_attnum_index | attnum</tt><br /><tt>
pg_inherits_relid_seqno_index | inhrelid</tt><br /><tt> pg_inherits_relid_seqno_index | inhseqno</tt><br
/><tt> pg_largeobject_loid_pn_index | loid</tt><br /><tt> pg_largeobject_loid_pn_index |
pageno</tt><br/><tt> pg_listener_pid_relname_index | listenerpid</tt><br /><tt>
pg_listener_pid_relname_index | relname</tt><br /><tt> pg_operator_oprname_l_r_k_index | oprname</tt><br
/><tt> pg_operator_oprname_l_r_k_index | oprleft</tt><br /><tt> pg_operator_oprname_l_r_k_index |
oprright</tt><br/><tt> pg_operator_oprname_l_r_k_index | oprkind</tt><br /><tt>
pg_proc_proname_narg_type_index| proname</tt><br /><tt> pg_proc_proname_narg_type_index | pronargs</tt><br
/><tt> pg_proc_proname_narg_type_index | proargtypes</tt><br /><tt> pg_statistic_relid_att_index |
starelid</tt><br/><tt> pg_statistic_relid_att_index | staattnum</tt><br /><tt>
primarytest2_pkey | col1</tt><br /><tt> primarytest2_pkey | col2</tt><br
/><tt> (29 rows)</tt><p>What do people think about this issue? It doesn't seem to have <br />major consequences
now,but it should probably be fixed? When <br />system tables have referential integrity enforced, then it will be an
<br/>issue. Can this be put on the TODO list? <p>Regards, <p>Joe Mitchell <br />Great Bridge LLC
