CUBRID Processes

Hi Artem,

Please see CUBRID Architecture: http://www.cubrid.org/manual/840/en/Structure

When starting CUBRID, depending on CUBRID version, the following processes are started:

 - cub_broker (probably 2 processes)

 - cub_cas (probably 10 processes)

 - cub_server (1 process)

 - cub_auto (1 process)

 - cub_master (1 process)

 - cub_js (1 process)

cub_broker and cub_cas

This processes belong to the CUBRID middle layer. An application which uses CUBRID connects to a cub_broker process which, in turn assigns a cub_cas process to it, which will handle application requests. How many cub_broker processes are started and how many cub_cas processes will be handled by each can be configured in cubrid_broker.conf (see http://www.cubrid.org/manual/840/en/cubrid_broker.conf). The default configuration starts two brokers with a minimum of 5 cub_cas processes each.

cub_server

The cub_server process is the actual storage engine. There will be a cub_server process for each database you start.

cub_auto, cub_js

I think this processes belong to the manager service. If you don't want to use CUBRID Manager, you can disable them by modifying cubrid.conf (see: http://www.cubrid.org/manual/840/en/cubrid.conf)

Modules started by default are configured in cubrid.conf:

# The list of processes to be started automatically by 'cubrid service start' command

# Any combinations are available with server, broker, manager and heartbeat.

service=server,broker,manager