Severe Delay in TICDC Data Collection

[translator_bot]

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

Original topic: TICDC采集延时严重

| username: TiDBer_Q9No79gQ

---

【TiDB Usage Environment】Issues occurred in both production and testing environments, but the following description pertains to the production environment.

【TiDB Version】5.7.25-TiDB-v5.4.0

【Background】
The production environment has over 6000 databases, each with 85 tables. After enabling TICDC collection, 51 merchants are collected, each with 15 tables, totaling 765 tables.

【Encountered Issues】
The collection delay is significant. When creating new tables in just 10 databases, the delay normalizes after 15 minutes. However, in the aforementioned production environment, delays of up to 1 day have been observed before synchronization returns to normal.

【Reproduction Path】
During the TICDC service process, an action involves deleting 2 tables from over 6000 databases, recreating these 2 tables, and creating indexes on them. This type of operation consistently causes delays.

【Issue Phenomenon and Impact】

[translator_bot]

| username: HACK | Original post link

---

Is the amount of data in the table large? How many changefeeds have been created?

[translator_bot]

| username: TiDBer_Q9No79gQ | Original post link

---

Some tables have data volumes exceeding ten million, but there aren’t many such tables.

[translator_bot]

| username: tidb狂热爱好者 | Original post link

---

This is a batch DDL. Execute one DDL every second.

[translator_bot]

| username: TiDBer_Q9No79gQ | Original post link

---

What is causing the delay? The ticdc service shouldn’t have much impact on the overall TiDB, right?

[translator_bot]

| username: 箱子NvN | Original post link

---

The official documentation has this. You can refer to it.

[translator_bot]

| username: system | Original post link

---

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.