Recent RI changes have broken something
От | Oliver Elphick |
---|---|
Тема | Recent RI changes have broken something |
Дата | |
Msg-id | 200001312218.WAA32035@linda.lfix.co.uk обсуждение исходный текст |
Ответы |
Re: [HACKERS] Recent RI changes have broken something
|
Список | pgsql-hackers |
Does anyone know what is happening here? There is no other user; the CREATE TABLE is part of a much larger script that was working OK last Thursday. The script drops and recreates an entire database. bray=# \d country Table "country"Attribute | Type | Modifier -----------+---------+----------id | char(2) | not nullname | text | not nullregion | text | telcode | text | Index: country_pkey Constraint: (id ~ '[A-Z]{2}'::text) bray=# create table country_ccy ( country char(2) references country (id) match full, ccy char(3) references currency (symbol) match full, primary key (country, ccy) ) ; NOTICE: CREATE TABLE/PRIMARY KEY will create implicit index 'country_ccy_pkey' for table 'country_ccy' NOTICE: CREATE TABLE will create implicit trigger(s) for FOREIGN KEY check(s) ERROR: RelationClearRelation: relation 21645 modified while in use bray=# select relname from pg_class where oid = 21645;relname ---------country (1 row) country was referenced in a previous table's foreign key. In view of the notes on RelationClearRelation, I am wondering if refcount wasn't decremented after that table was created. -- Oliver Elphick Oliver.Elphick@lfix.co.uk Isle of Wight http://www.lfix.co.uk/oliver PGP key from public servers; key ID32B8FAA1 ======================================== "My son, if sinners entice thee, consent thou not." Proverbs 1:10
В списке pgsql-hackers по дате отправления: