Production Deployment from Binary Tarball

This guide provides installation instructions from a binary tarball on Linux. A complete TiDB cluster contains PD, TiKV, and TiDB. To start the database service, follow the order of PD -> TiKV -> TiDB. To stop the database service, follow the order of stopping TiDB -> TiKV -> PD.

See also local deployment and testing environment deployment.

Prepare

Before you start, see TiDB architecture and Software and Hardware Recommendations. Make sure the following requirements are satisfied:

Operating system

For the operating system, it is recommended to use RHEL/CentOS 7.3 or higher. The following additional requirements are recommended:

Network and firewall

Operating system parameters

Database running user settings

TiDB components and default ports

Before you deploy a TiDB cluster, see the required components and optional components.

TiDB database components (required)

See the following table for the default ports for the TiDB components:

TiDB database components (optional)

See the following table for the default ports for the optional TiDB components:

Create a database running user account

1. Log in to the machine using the root user account and create a database running user account (tidb) using the following command:

   # useradd tidb -m

2. Switch the user from root to tidb by using the following command. You can use this tidb user account to deploy your TiDB cluster.

   # su - tidb

Download the official binary package

# Download the package.
$ wget https://download.pingcap.org/tidb-latest-linux-amd64.tar.gz
$ wget https://download.pingcap.org/tidb-latest-linux-amd64.sha256

# Check the file integrity. If the result is OK, the file is correct.
$ sha256sum -c tidb-latest-linux-amd64.sha256

# Extract the package.
$ tar -xzf tidb-latest-linux-amd64.tar.gz
$ cd tidb-latest-linux-amd64

Multiple nodes cluster deployment

For the production environment, multiple nodes cluster deployment is recommended. Before you begin, see Software and Hardware Recommendations.

Assuming that you have six nodes, you can deploy 3 PD instances, 3 TiKV instances, and 1 TiDB instance. See the following table for details:

Follow the steps below to start PD, TiKV, and TiDB:

1. Start PD on Node1, Node2, and Node3 in sequence.

   $ ./bin/pd-server --name=pd1 \
                   --data-dir=pd \
                   --client-urls="http://192.168.199.113:2379" \
                   --peer-urls="http://192.168.199.113:2380" \
                   --initial-cluster="pd1=http://192.168.199.113:2380,pd2=http://192.168.199.114:2380,pd3=http://192.168.199.115:2380" \
                   -L "info" \
                   --log-file=pd.log &
   
   $ ./bin/pd-server --name=pd2 \
                   --data-dir=pd \
                   --client-urls="http://192.168.199.114:2379" \
                   --peer-urls="http://192.168.199.114:2380" \
                   --initial-cluster="pd1=http://192.168.199.113:2380,pd2=http://192.168.199.114:2380,pd3=http://192.168.199.115:2380" \
                   -L "info" \
                   --log-file=pd.log &
   
   $ ./bin/pd-server --name=pd3 \
                   --data-dir=pd \
                   --client-urls="http://192.168.199.115:2379" \
                   --peer-urls="http://192.168.199.115:2380" \
                   --initial-cluster="pd1=http://192.168.199.113:2380,pd2=http://192.168.199.114:2380,pd3=http://192.168.199.115:2380" \
                   -L "info" \
                   --log-file=pd.log &

2. Start TiKV on Node4, Node5 and Node6.

   $ ./bin/tikv-server --pd="192.168.199.113:2379,192.168.199.114:2379,192.168.199.115:2379" \
                     --addr="192.168.199.116:20160" \
                     --status-addr="192.168.199.116:20180" \
                     --data-dir=tikv \
                     --log-file=tikv.log &
   
   $ ./bin/tikv-server --pd="192.168.199.113:2379,192.168.199.114:2379,192.168.199.115:2379" \
                     --addr="192.168.199.117:20160" \
                     --status-addr="192.168.199.117:20180" \
                     --data-dir=tikv \
                     --log-file=tikv.log &
   
   $ ./bin/tikv-server --pd="192.168.199.113:2379,192.168.199.114:2379,192.168.199.115:2379" \
                     --addr="192.168.199.118:20160" \
                     --status-addr="192.168.199.118:20180" \
                     --data-dir=tikv \
                     --log-file=tikv.log &

3. Start TiDB on Node1.

   $ ./bin/tidb-server --store=tikv \
                     --path="192.168.199.113:2379,192.168.199.114:2379,192.168.199.115:2379" \
                     --log-file=tidb.log &

4. Use the MySQL client to connect to TiDB.

   $ mysql -h 192.168.199.113 -P 4000 -u root -D test

Note:

• If you start TiKV or deploy PD in the production environment, it is highly recommended to specify the path for the configuration file using the --config parameter. If the parameter is not set, TiKV or PD does not read the configuration file.
• To tune TiKV, see Performance Tuning for TiKV.
• If you use nohup to start the cluster in the production environment, write the startup commands in a script and then run the script. If not, the nohup process might abort because it receives exceptions when the Shell command exits. For more information, see The TiDB/TiKV/PD process aborts unexpectedly.

For the deployment and use of TiDB monitoring services, see Deploy Monitoring Services for the TiDB Cluster and TiDB Monitoring API.