In this lesson we will:
- Use the command line interface (CLI) to start the ClickHouse server and then connect to it from a local client.
Starting The Server
The training virtual machine already has the ClickHouse binary installed, but the server has not been started. We can begin therefore by simply running the clickhouse server command to start the process.
cd clickhouse
./clickhouse server
This will output a fairly lengthy set of logs, but a number of informational messages show that the server is starting cleanly:
2021.11.25 12:17:42.942929 [ 25 ] {} <Information> : Starting ClickHouse 21.11.4.14 with revision 54456, build id: 8332E80F7ACF6DE1033E1A6D9F16415C6E03DBF0, PID 25 2021.11.25 12:17:42.946195 [ 25 ] {} <Information> Application: starting up 2021.11.25 12:17:42.946278 [ 25 ] {} <Information> Application: OS name: Linux, version: 5.10.47-linuxkit, architecture: x86_64
2021.11.25 12:17:45.423613 [ 25 ] {} <Information> Application: Calculated checksum of the binary: 805E158CCA2B20788668232443B9CEE6, integrity check passed.
2021.11.25 12:17:45.555272 [ 25 ] {} <Information> DatabaseCatalog: Found 0 partially dropped tables. Will load them and retry removal. 2021.11.25 12:17:45.561421 [ 25 ] {} <Information> DatabaseAtomic (default): Metadata processed, database default has 0 tables and 0 dictionaries in total. 2021.11.25 12:17:45.561494 [ 25 ] {} <Information> TablesLoader: Parsed metadata of 0 tables in 1 databases in 0.000178457 sec
2021.11.25 12:17:45.561503 [ 25 ] {} <Information> TablesLoader: Loading 0 tables with 0 dependency level 2021.11.25 12:17:45.561511 [ 25 ] {} <Information> DatabaseAtomic (default): Starting up tables. 2021.11.25 12:17:45.561518 [ 25 ] {} <Information> DatabaseAtomic (system): Starting up tables.
2021.11.25 12:17:45.565785 [ 25 ] {} <Information> BackgroundSchedulePool/BgSchPool: Create BackgroundSchedulePool with 128 threads
Starting The Client
In a second shell, we can then start the ClickHouse client command. This is a command line tool which allows us to send commands to the server which we've just started:
./clickhouse client
The connection should be succesful out of the box. This is because the client assumes some defaults for the connection, such as the fact that we would like to connect to localhost, with the username default and a blank password.
If we were connecting to a different secured ClickHouse machine or with a different username or password, we can specify these details at the command line:
clickhouse-client -i clickhouse.timeflow.com:2131 -u benjaminwootton --password ABC123
For the purpose of the lesson, the defaults are fine so the client should connect succesfully. We can then issue a command within the ClickHouse client to check connectivity:
show databases
Outputs:
SHOW DATABASES
Query id: a3f1db00-87af-47a8-9096-dad14df00589
┌─name───────────────┐
│ INFORMATION_SCHEMA │
│ default │
│ information_schema │
│ system │
└────────────────────┘
4 rows in set. Elapsed: 0.004 sec.
This shows that the server is running, and we now have connectivity between our client the server.