Ticdc Bidirectional Replication Causes Data Desynchronization When Creating New Tables

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

Original topic: Ticdc双向复制创建新表数据不同步

| username: TiDBer_hwEZA4rV

TiDB&Ticdc version 6.5.1
Data Center A
Cluster 1
Cluster 2
Data Center B
Cluster 3

Cluster 1 and Cluster 3 bidirectional replication
Cluster 1 unidirectional synchronization to Cluster 2

Cluster 1 and 3 create table X (bidirectional replication DDL not synchronized)
Data inserted into Cluster 1 is not synchronized to Cluster 3, but can be synchronized to Cluster 2
How can this issue be resolved?

| username: Billmay表妹 | Original post link

Based on your description, the bidirectional replication between Cluster 1 and Cluster 3 did not synchronize the DDL of table X, which may cause data to not be correctly synchronized between Cluster 1 and Cluster 3. To resolve this issue, you can try the following steps:

  1. Manually execute the DDL of table X in both Cluster 1 and Cluster 3 to ensure that the table structure is consistent in both clusters. You can use the CREATE TABLE or ALTER TABLE commands to create or modify the table structure.
  2. Confirm whether the TiCDC configuration is correct. TiCDC is a component in TiDB used for data change capture and synchronization. If the TiCDC configuration is incorrect, it may cause data synchronization to fail. You can refer to the relevant sections in the official TiDB documentation [1] to check whether the TiCDC configuration is correct.
  3. Confirm whether the TiCDC status is normal. You can use the cdc cli changefeed query command to query the TiCDC status and check for any anomalies. If the TiCDC status is abnormal, you can try restarting TiCDC or recreating the TiCDC changefeed.
  4. Confirm whether the TiKV status is normal. If the TiKV status is abnormal, it may cause data synchronization to fail. You can use the monitoring panel in the TiDB Dashboard to check the TiKV status and look for any anomalies. If the TiKV status is abnormal, you can try restarting TiKV or troubleshooting the issue.

If you still cannot resolve the problem, it is recommended that you provide more detailed information, such as the TiCDC configuration, TiKV status, etc., so that we can better assist you in resolving the issue.

| username: Fly-bird | Original post link

Is there any network latency between Cluster 1 and Cluster 3?

| username: nongfushanquan | Original post link

Check whether the changefeed from Cluster 1 to Cluster 3 has filter rules configured, and if so, whether the rules include the newly created table.

| username: asddongmen | Original post link

Hello, could you please provide a topology diagram for analysis?

| username: ajin0514 | Original post link

It could be due to network issues.