Help needed: What tool is suitable for data migration within a TiDB cluster?

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

Original topic: 求助大佬 tidb集群内数据迁移用什么工具比较合适呀

| username: TiDBer_GQ1Pmqa3

[TiDB Usage Environment] Production Environment / Testing / PoC
[TiDB Version]
[Reproduction Path] What operations were performed when the issue occurred
[Encountered Issue: Issue Phenomenon and Impact]
[Resource Configuration]
[Attachments: Screenshots / Logs / Monitoring]

After synchronizing MySQL data to the local TiDB (db1 database) cluster through TiDM, I want to continue monitoring this db1 database in TiDB and select some tables to synchronize to another database in TiDB (db2). Which tool is more suitable for this?
Due to some reasons, the leader has requested not to start another TiDM task to monitor MySQL.

| username: 啦啦啦啦啦 | Original post link

TiCDC should be able to meet the requirements.

| username: 像风一样的男子 | Original post link

Are you synchronizing db1 to db2 within the same cluster?

| username: TiDBer_GQ1Pmqa3 | Original post link

Yes, you can perform full and incremental synchronization on individual tables.

| username: TiDBer_GQ1Pmqa3 | Original post link

TiCDC can only synchronize incremental data, right?

| username: 啦啦啦啦啦 | Original post link

Yes, you need to first synchronize a full copy of the data and then use TiCDC to synchronize the incremental data.

| username: TiDBer_GQ1Pmqa3 | Original post link

Okay, boss, I’ll check the documentation again. How do I perform a full synchronization?

| username: 啦啦啦啦啦 | Original post link

Just use Dumpling.

| username: TiDBer_GQ1Pmqa3 | Original post link

Got it! Thanks, boss!

| username: 像风一样的男子 | Original post link

It seems that TiDB’s own components do not support it, but third-party tools like FlinkCDC can achieve it.

| username: 像风一样的男子 | Original post link

Dumpling cannot monitor data changes in real-time for incremental synchronization, right?

| username: 啦啦啦啦啦 | Original post link

First, use Dumpling to import the full data of these tables, and then use TiCDC to add the incremental data. I haven’t tried the situation where both the upstream and downstream are the same TiDB, but it should be fine.

| username: TiDBer_GQ1Pmqa3 | Original post link

Okay, let me give it a try.

| username: 像风一样的男子 | Original post link

I haven’t seen this scenario before. Theoretically, it should be feasible.

| username: TiDBer_GQ1Pmqa3 | Original post link

I have tried Flink CDC, it requires TiDB to enable binlog, right?

| username: xfworld | Original post link

7.3.0 is not LTS… it’s a short-term version (DMR), it might not have upgrade support, so this needs to be noted.

| username: TiDBer_GQ1Pmqa3 | Original post link

Got it!

| username: xfworld | Original post link

FlinkCDC is a connector for Flink, and it is a completely different solution from TiCDC. I recommend you use TiCDC… to avoid pitfalls…

| username: Kongdom | Original post link

TiCDC is definitely the top choice~

| username: Fly-bird | Original post link

dumpling+cdc