- About DM
- Usage Scenarios
- Quick Start
- Manage Migration Tasks
- Performance Tuning
- Release Notes
This document introduces how to maintain a DM cluster using the TiUP DM component. For the complete steps of DM deployment, refer to Deploy a DM Cluster Using TiUP.
- Make sure that the ports among the following components are interconnected
8291by default) among the DM-master nodes are interconnected.
- Each DM-master node can connect to the
portof all DM-worker nodes (
- Each DM-worker node can connect to the
portof all DM-master nodes (
- The TiUP nodes can connect to the
portof all DM-master nodes (
- The TiUP nodes can connect to the
portof all DM-worker nodes (
For the help information of the TiUP DM component, run the following command:
tiup dm --help
Deploy a DM cluster for production Usage: tiup dm [flags] tiup dm [command] Available Commands: deploy Deploy a DM cluster for production start Start a DM cluster stop Stop a DM cluster restart Restart a DM cluster list List all clusters destroy Destroy a specified DM cluster audit Show audit log of cluster operation exec Run shell command on host in the dm cluster edit-config Edit DM cluster config display Display information of a DM cluster reload Reload a DM cluster's config and restart if needed upgrade Upgrade a specified DM cluster patch Replace the remote package with a specified package and restart the service scale-out Scale out a DM cluster scale-in Scale in a DM cluster import Import an exist DM 1.0 cluster from dm-ansible and re-deploy 2.0 version help Help about any command Flags: -h, --help help for tiup-dm --native-ssh Use the native SSH client installed on local system instead of the build-in one. --ssh-timeout int Timeout in seconds to connect host via SSH, ignored for operations that don't need an SSH connection. (default 5) -v, --version version for tiup-dm --wait-timeout int Timeout in seconds to wait for an operation to complete, ignored for operations that don't fit. (default 60) -y, --yes Skip all confirmations and assumes 'yes'
After the cluster is successfully deployed, view the cluster list by running the following command:
tiup dm list
Name User Version Path PrivateKey ---- ---- ------- ---- ---------- prod-cluster tidb v2.0.0 /root/.tiup/storage/dm/clusters/test /root/.tiup/storage/dm/clusters/test/ssh/id_rsa
After the cluster is successfully deployed, start the cluster by running the following command:
tiup dm start prod-cluster
If you forget the name of your cluster, view the cluster list by running
tiup dm list.
TiUP provides the
tiup dm display command to view the status of each component in the cluster. With this command, you do not have to log in to each machine to see the component status. The usage of the command is as follows:
tiup dm display prod-cluster
dm Cluster: prod-cluster dm Version: v2.0.0 ID Role Host Ports OS/Arch Status Data Dir Deploy Dir -- ---- ---- ----- ------- ------ -------- ---------- 172.19.0.101:9093 alertmanager 172.19.0.101 9093/9094 linux/x86_64 Up /home/tidb/data/alertmanager-9093 /home/tidb/deploy/alertmanager-9093 172.19.0.101:8261 dm-master 172.19.0.101 8261/8291 linux/x86_64 Healthy|L /home/tidb/data/dm-master-8261 /home/tidb/deploy/dm-master-8261 172.19.0.102:8261 dm-master 172.19.0.102 8261/8291 linux/x86_64 Healthy /home/tidb/data/dm-master-8261 /home/tidb/deploy/dm-master-8261 172.19.0.103:8261 dm-master 172.19.0.103 8261/8291 linux/x86_64 Healthy /home/tidb/data/dm-master-8261 /home/tidb/deploy/dm-master-8261 172.19.0.101:8262 dm-worker 172.19.0.101 8262 linux/x86_64 Free /home/tidb/data/dm-worker-8262 /home/tidb/deploy/dm-worker-8262 172.19.0.102:8262 dm-worker 172.19.0.102 8262 linux/x86_64 Free /home/tidb/data/dm-worker-8262 /home/tidb/deploy/dm-worker-8262 172.19.0.103:8262 dm-worker 172.19.0.103 8262 linux/x86_64 Free /home/tidb/data/dm-worker-8262 /home/tidb/deploy/dm-worker-8262 172.19.0.101:3000 grafana 172.19.0.101 3000 linux/x86_64 Up - /home/tidb/deploy/grafana-3000 172.19.0.101:9090 prometheus 172.19.0.101 9090 linux/x86_64 Up /home/tidb/data/prometheus-9090 /home/tidb/deploy/prometheus-9090
Status column uses
Down to indicate whether the service is running normally.
For the DM-master component,
|L might be appended to a status, which indicates that the DM-master node is a Leader. For the DM-worker component,
Free indicates that the current DM-worker node is not bound to an upstream.
Scaling in a cluster means making some node(s) offline. This operation removes the specified node(s) from the cluster and deletes the remaining data files.
When you scale in a cluster, DM operations on DM-master and DM-worker components are performed in the following order:
- Stop component processes.
- Call the API for DM-master to delete the
- Clean up the data files related to the node.
The basic usage of the scale-in command:
tiup dm scale-in <cluster-name> -N <node-id>
To use this command, you need to specify at least two arguments: the cluster name and the node ID. The node ID can be obtained by using the
tiup dm display command in the previous section.
For example, to scale in the DM-worker node on
172.16.5.140 (similar to scaling in DM-master), run the following command:
tiup dm scale-in prod-cluster -N 172.16.5.140:8262
The scale-out operation has an inner logic similar to that of deployment: the TiUP DM component first ensures the SSH connection of the node, creates the required directories on the target node, then executes the deployment operation, and starts the node service.
For example, to scale out a DM-worker node in the
prod-cluster cluster, take the following steps (scaling out DM-master has similar steps):
scale.yamlfile and add information of the new worker node:
You need to create a topology file, which includes only the description of the new nodes, not the existing nodes. For more configuration items (such as the deployment directory), refer to this TiUP configuration parameter example.
--- worker_servers: - host: 172.16.5.140
Perform the scale-out operation. TiUP DM adds the corresponding nodes to the cluster according to the port, directory, and other information described in
tiup dm scale-out prod-cluster scale.yaml
After the command is executed, you can check the status of the scaled-out cluster by running
tiup dm display prod-cluster.
The rolling upgrade process is made as transparent as possible to the application, and does not affect the business. The operations vary with different nodes.
You can run the
tiup dm upgrade command to upgrade a DM cluster. For example, the following command upgrades the cluster to v2.0.1:
tiup dm upgrade prod-cluster v2.0.1
If you want to dynamically update the component configurations, the TiUP DM component saves a current configuration for each cluster. To edit this configuration, execute the
tiup dm edit-config <cluster-name> command. For example:
tiup dm edit-config prod-cluster
TiUP DM opens the configuration file in the vi editor. If you want to use other editors, use the
EDITOR environment variable to customize the editor, such as
export EDITOR=nano. After editing the file, save the changes. To apply the new configuration to the cluster, execute the following command:
tiup dm reload prod-cluster
The command sends the configuration to the target machine and restarts the cluster to make the configuration take effect.
For normal upgrade, you can use the
upgrade command. But in some scenarios, such as debugging, you might need to replace the currently running component with a temporary package. To achieve this, use the
tiup dm patch --help
Replace the remote package with a specified package and restart the service Usage: tiup dm patch <cluster-name> <package-path> [flags] Flags: -h, --help help for patch -N, --node strings Specify the nodes --overwrite Use this package in the future scale-out operations -R, --role strings Specify the role --transfer-timeout int Timeout in seconds when transferring dm-master leaders (default 300) Global Flags: --native-ssh Use the native SSH client installed on local system instead of the build-in one. --ssh-timeout int Timeout in seconds to connect host via SSH, ignored for operations that don't need an SSH connection. (default 5) --wait-timeout int Timeout in seconds to wait for an operation to complete, ignored for operations that don't fit. (default 60) -y, --yes Skip all confirmations and assumes 'yes'
If a DM-master hotfix package is in
/tmp/dm-master-hotfix.tar.gz and you want to replace all the DM-master packages in the cluster, run the following command:
tiup dm patch prod-cluster /tmp/dm-master-hotfix.tar.gz -R dm-master
You can also replace only one DM-master package in the cluster:
tiup dm patch prod-cluster /tmp/dm--hotfix.tar.gz -N 172.16.4.5:8261
- TiUP does not support importing the DM Portal component in a DM 1.0 cluster.
- You need to stop the original cluster before importing.
- TiUP only supports importing to a DM cluster of v2.0.0-rc.2 or a later version.
- The deployment directories of some components are different from those of the original cluster. You can execute the
displaycommand to view the details.
tiup update --self && tiup update dmbefore importing to make sure that the TiUP DM component is the latest version.
- Only one DM-master node exists in the cluster after importing. Refer to Scale out a cluster to scale out the DM-master.
Before TiUP is released, DM-Ansible is often used to deploy DM clusters. To enable TiUP to take over the DM 1.0 cluster deployed by DM-Ansible, use the
For example, to import a cluster deployed using DM Ansible:
tiup dm import --dir=/path/to/dm-ansible --cluster-version v2.0.0-rc.2
tiup list dm-master to view the latest cluster version supported by TiUP.
The process of using the
import command is as follows:
- TiUP generates a topology file
topology.ymlbased on the DM cluster previously deployed using DM-Ansible.
- After confirming that the topology file has been generated, you can use it to deploy the DM cluster of v2.0 or later versions.
After the deployment is completed, you can execute the
tiup dm start command to start the cluster and begin the process of upgrading the DM kernel.
To view the operation log, use the
audit command. The usage of the
audit command is as follows:
Usage: tiup dm audit [audit-id] [flags] Flags: -h, --help help for audit
[audit-id] argument is not specified, the command shows a list of commands that have been executed. For example:
tiup dm audit
ID Time Command -- ---- ------- 4D5kQY 2020-08-13T05:38:19Z tiup dm display test 4D5kNv 2020-08-13T05:36:13Z tiup dm list 4D5kNr 2020-08-13T05:36:10Z tiup dm deploy -p prod-cluster v2.0.0 ./examples/dm/minimal.yaml
The first column is
audit-id. To view the execution log of a certain command, pass the
audit-id argument as follows:
tiup dm audit 4D5kQY
To run commands on a host in the DM cluster, use the
exec command. The usage of the
exec command is as follows:
Usage: tiup dm exec <cluster-name> [flags] Flags: --command string the command run on cluster host (default "ls") -h, --help help for exec -N, --node strings Only exec on host with specified nodes -R, --role strings Only exec on host with specified roles --sudo use root permissions (default false)
For example, to execute
ls /tmp on all DM nodes, run the following command:
tiup dm exec prod-cluster --command='ls /tmp'
TiUP integrates the DM cluster controller
Run the following command to use dmctl:
tiup dmctl [args]
Specify the version of dmctl:
tiup dmctl:v2.0.0 [args]
The previous dmctl command to add a source is
dmctl --master-addr master1:8261 operate-source create /tmp/source1.yml. After dmctl is integrated into TiUP, the command is:
tiup dmctl --master-addr master1:8261 operate-source create /tmp/source1.yml
All operations above performed on the cluster machine use the SSH client embedded in TiUP to connect to the cluster and execute commands. However, in some scenarios, you might also need to use the SSH client native to the control machine system to perform such cluster operations. For example:
- To use a SSH plug-in for authentication
- To use a customized SSH client
Then you can use the
--native-ssh command-line flag to enable the system-native command-line tool:
- Deploy a cluster:
tiup dm deploy <cluster-name> <version> <topo> --native-ssh
- Start a cluster:
tiup dm start <cluster-name> --native-ssh
- Upgrade a cluster:
tiup dm upgrade ... --native-ssh
You can add
--native-ssh in all cluster operation commands above to use the system's native SSH client.
To avoid adding such a flag in every command, you can use the
TIUP_NATIVE_SSH system variable to specify whether to use the local SSH client:
export TIUP_NATIVE_SSH=true # or export TIUP_NATIVE_SSH=1 # or export TIUP_NATIVE_SSH=enable
If you specify this environment variable and
--native-ssh at the same time,
--native-ssh has higher priority.
During the process of cluster deployment, if you need to use a password for connection or
passphraseis configured in the key file, you must ensure that
sshpassis installed on the control machine; otherwise, a timeout error is reported.
- Maintain a DM Cluster Using TiUP
- View the cluster list
- Start the cluster
- Check the cluster status
- Scale in a cluster
- Scale out a cluster
- Rolling upgrade
- Update configuration
- Update component
- Import and upgrade a DM 1.0 cluster deployed using DM-Ansible
- View the operation log
- Run commands on a host in the DM cluster
- Use the system's native SSH client to connect to cluster