Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.
Original topic: DM: 同步任务上游的IP发生了变化，是否可以重新配置直接start继续增量同步？ （还是原来的实例，只是ip没变）
As mentioned,
the upstream instance needs a network migration, and the IP will change. The overall binlog remains unchanged.
In this case, can the task that is already in the full → incremental phase be directly stopped, then modify the upstream data source configuration, and restart?
I understand that there should be no problem.
After stopping the task, it cannot be restarted.
You can try the following steps:
- pause-task
- operate-source update
- query-status (confirm whether the IP of the data source in the task has changed)
- resume-task
It should be possible, just make sure the source-id doesn’t change.
“Stop means to halt, and after stopping, you can’t resume the previous task upon restarting. You should pause the task instead.”
The task name hasn’t changed; there is DM metadata within TiDB. It will restart, right?
Directly starting it will result in an error; you need to modify the task metadata information or delete the metadatabase table.
I’m a bit confused about this part. I’ve actually performed this operation many times. I stopped a normal task, modified the configuration file (filter rules), and then started the task (without --remove-meta). It started normally, and I didn’t see any issues with the synchronization. 
What version are you using? I haven’t been keeping up with the iterations of new versions, it might be a new feature in the later versions.
What command is used to stop the task?
I’m still using the old version, 2.0.
|
username: 超7成网友
|
username: 
|
username: 
|
username: