Multiple tasks share a single worker under DM

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: DM下多个task共用一个worker

| username: 特雷西-迈克-格雷迪

[TiDB Usage Environment] Production Environment
[TiDB Version] v5.2.0
[Reproduction Path]
[Encountered Problem: Problem Phenomenon and Impact]
DM: Version 2.0.6
Multiple tasks are using the same worker. Generally, when starting a task, the new task will bind to a new worker, but our DM cluster is now a bit strange, with multiple tasks sharing the same worker:
task_1
“worker”: “dm-192.122.22.16-8262”,
task_2
“worker”: “dm-192.122.22.24-18265”
task_3
“worker”: “dm-192.122.22.24-18265”
task_4
“worker”: “dm-192.122.22.24-18265”,
task_5
“worker”: “dm-192.122.22.17-8262”,
task_6
“worker”: “dm-192.122.22.17-8262”,
task_7
“worker”: “dm-192.122.22.17-8262”

| username: Hacker_7p5BxP3A | Original post link

Take a screenshot and display the details of the worker and task.

| username: 特雷西-迈克-格雷迪 | Original post link

I can’t upload pictures, not sure why.

| username: 特雷西-迈克-格雷迪 | Original post link

The query-status shows that there are three different tasks, and their “worker”: “dm-xxxx-8262” is the same.

| username: Hacker_7p5BxP3A | Original post link

Are the data sources bound to these tasks the same?

| username: 特雷西-迈克-格雷迪 | Original post link

It’s the same.

| username: 有猫万事足 | Original post link

If the data source bound to the task is the same, it is naturally the same DM worker. To be precise, the data source is bound to the DM worker.

For example, if dm1 is connected to mysql1, as long as the data source set in the task is mysql1, it will be submitted to dm1 for execution.