Thread: Avoiding RI failure with INSERT ... SELECT FROM
Hello! I got an unexpected SQL failure last night, and want to see how
to prevent it in the future. This is with 8.1.3.
I have a couple of insert statements that run in a transaction block,
initializing rows that will be used the next day:
INSERT into item_hit_log SELECT item_id, CURRENT_DATE + CAST('1 day' AS interval), 0 FROM items where
item_state= 'available';
INSERT into item_view_log SELECT item_id, CURRENT_DATE + CAST('1 day' AS interval), 0 FROM items
whereitem_state = 'available';
The "items" table has a few hundred thousand rows in it, so this takes a
bit a run.
The error we got last night was:
ERROR: insert or update on table "item_hit_log" violates foreign key
constraint "item_id_fk" DETAIL: Key (item_id)=(451226) is not present in table "items".
Re-running the transaction block a few minutes later worked.
What's an alternate design that avoids this possible error?
Thanks!
Mark
On 7/12/06, Mark Stosberg <mark@summersault.com> wrote:
I would recommend you list your values in your insert statements:
INSERT into item_hit_log (
item_id_fk, ....
)
SELECT ....
That makes it less prone to problems in the future (like if the column orders change) and makes it easier for others to understand and help you with.
Does the items table allow deletes? If so, your insert may be attempting to do an insert for an item_id that was deleted after the select and before the insert. Don't know if PostgreSQL will prevent that with table locking or not.
==================================================================
Aaron Bono
Aranya Software Technologies, Inc.
http://www.aranya.com
==================================================================
Hello! I got an unexpected SQL failure last night, and want to see how
to prevent it in the future. This is with 8.1.3.
I have a couple of insert statements that run in a transaction block,
initializing rows that will be used the next day:
INSERT into item_hit_log
SELECT item_id, CURRENT_DATE + CAST('1 day' AS interval), 0
FROM items where item_state = 'available';
INSERT into item_view_log
SELECT item_id, CURRENT_DATE + CAST('1 day' AS interval), 0
FROM items where item_state = 'available';
I would recommend you list your values in your insert statements:
INSERT into item_hit_log (
item_id_fk, ....
)
SELECT ....
That makes it less prone to problems in the future (like if the column orders change) and makes it easier for others to understand and help you with.
The "items" table has a few hundred thousand rows in it, so this takes a
bit a run.
The error we got last night was:
ERROR: insert or update on table "item_hit_log" violates foreign key
constraint "item_id_fk"
DETAIL: Key (item_id)=(451226) is not present in table "items".
Re-running the transaction block a few minutes later worked.
What's an alternate design that avoids this possible error?
Does the items table allow deletes? If so, your insert may be attempting to do an insert for an item_id that was deleted after the select and before the insert. Don't know if PostgreSQL will prevent that with table locking or not.
==================================================================
Aaron Bono
Aranya Software Technologies, Inc.
http://www.aranya.com
==================================================================
Hello,
This is a re-phrasing of question I asked recently, inquiring about a
an approach I'm considering.
My query is:
> INSERT into item_hit_log (item_id, hit_date, hit_count)> SELECT item_id, CURRENT_DATE + CAST('1 day' AS interval),
0> FROM items where item_state = 'available';
The "items" table has a few hundred thousand rows in it, and is likely
to be updated by other queries while this runs.
The error we got last night was:
ERROR: insert or update on table "item_hit_log" violates foreign key
constraint "item_id_fk" DETAIL: Key (item_id)=(451226) is not present in table "items".
Re-running the transaction block a few minutes later worked.
From reading the docs, it sounds like "SELECT ... FOR UPDATE"
might be the perfect solution here.
http://www.postgresql.org/docs/8.1/static/sql-select.html#SQL-FOR-UPDATE-SHARE
Since it's hard to test for this kind of failure, I wanted to get the
opinions of others here if this would be a good idea.
Thanks!
Mark