We use pg_repack to reorganize some large tables(between 100GB and 1TB). We don't have any locking issues while creating the triggers on the source table and swapping tables at the end. But every index creation on the shadow table causes locking issues. The lock waits start to happen just after the CREATE INDEX command starts on the shadow table. It does not last along the whole creation process. The locking disappears after a while, i.e it happens in the first 30 seconds of index creation, but the index creation takes 10 minutes.
Here is the command we are using:
pg_repack -k -e -h <host_name> -d <db_name> -U <user_name> -t <table_name> -o <column_list>
We are on pg_repack-1.4.6 and PostgreSQL 13.13.
We use
pg_repackto reorganize some large tables(between 100GB and 1TB). We don't have any locking issues while creating the triggers on the source table and swapping tables at the end. But every index creation on the shadow table causes locking issues. The lock waits start to happen just after theCREATE INDEXcommand starts on the shadow table. It does not last along the whole creation process. The locking disappears after a while, i.e it happens in the first 30 seconds of index creation, but the index creation takes 10 minutes.Here is the command we are using:
We are on
pg_repack-1.4.6andPostgreSQL 13.13.