Hey, I’ve been investigating SQL_BIG_SELECTS, but the MySQL documentation so far has been pretty unhelpful. I’m looking for some insight as to preventing errors like the one below from appearing.
ERROR 1104: The SELECT would examine too many records and probably take a very long time.
Check your WHERE and use SET OPTION SQL_BIG_SELECTS=1 if the SELECT is ok
- At how many rows does MySQL decide that a query is a “BIG SELECT”?
- Will proper indexing usually solve this issue?
- Is SQL_BIG_SELECTS considered a “last resort”, or is it good practice?
- How would someone set “SQL_BIG_SELECTS=1” in configuration (without having to execute the query)?
- Are there any other alternatives worth knowing?
Thanks in advance!
MySQL determines whether or not a query is a ‘big select’ based on the value of ‘max_join_size’. If the query is likely to have to examine more than this number of rows, it will consider it a ‘big select’. Use ‘show variables’ to view the value of the max join size.
I believe that indexing and particular a good where clause will prevent this problem from occuring.
SQL_BIG_SELECTS is used to prevent users from accidentally executing excessively large queries. It is okay to set it to ON in mysql.cnf or using the command-line option at startup.
You can set SQL_BIG_SELECTS in my.cnf or at server startup. It can also be set on a session basis with
SET SESSION SQL_BIG_SELECTS=1.Not that I can think of. I would just check your query to make sure that you really need to use it. Our servers have it turned on by default, and max_join_size is very large.