Commands

Commands are non SQL instructions

defragmentation

The CELL_* files in internal storage needs defragmentation, defrag is done in offline mode during the save process

By default a script execution will stop on first error

Sometimes this is not what you want (common case is multiple execution of a create statement)

To continue on all errors

continue_on_error

To stop on all errors

stop_on_error

To continue on some errors

continue_on_error(err_code1, err_code2, ...)

start.sh
stop.sh
bounce.sh
show_log.sh

Exports data_model and/or data to a directory

./export.sh <export_dir> <data_model y/n> <data y/n> <nb_thread> <dry_run y/n>

Import data_model and/or data from an export directory

./import.sh <export_dir> <data_model y/n> <data y/n> <dry_run y/n>

From 02_sql

./sql.sh the_user/the_password@sb_host:sb_port [--file:<file.sql> --i]

i means interactive (the file is executed but you must press enter after each execution)

From 01_database (connection to local SB)

./sql.sh [--file:<file.sql> --i]

  Note about logging: sql.sh will create a log file (sql.log), this file is bit different from stdout, it is used to generate this documentation !!

n tells sql.sh to "read but not run"
y tells sql.sh to restart normally

something I want to see in sql.log without being executed
it can be on several lines but cannot contain a semi column

same as bounce.sh without killing the process

same as stop.sh without killing the process