You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
currently the estimate_count function calculates table rows without including the custom --where option provided by the user
example: if a table has id column from 1 to 1000 and we want to dump only from 100 to 200, estimates are calculated for the full table. This will have no impact on small tables but can cause serious inefficiencies if the tables have hundreds of millions of rows.
We dump tables with --where clause and we noticed that job count in the queue is the same no matter the range of columns we are dumping
To Reproduce
Command executed:
run a mydumper for any table with primary key
What mydumper and myloader version has been used? mydumper 0.12.1, built against MySQL 5.7.33-36
Expected behavior
rows are estimated including the --where clause
Environment (please complete the following information):
all OS types
The text was updated successfully, but these errors were encountered:
oh wait...including a where clause to the EXPLAIN SELECT is not gonna work because of this... this could be a feature to provide a way to allow users to dump tables for a specific range on a index
Describe the bug
--where
option provided by the userid
column from1 to 1000
and we want to dump only from100 to 200
, estimates are calculated for the full table. This will have no impact on small tables but can cause serious inefficiencies if the tables have hundreds of millions of rows.--where
clause and we noticed that job count in the queue is the same no matter the range of columns we are dumpingTo Reproduce
Command executed:
What mydumper and myloader version has been used?
mydumper 0.12.1, built against MySQL 5.7.33-36
Expected behavior
--where
clauseEnvironment (please complete the following information):
The text was updated successfully, but these errors were encountered: