Aiden comments

Results 6 comments of


                                            Aiden

[fix-issue-2676] repair a snapshot-split bug:

> Hi @AidenPerce, could you please rebase this PR with latest `master` branch before it could be merged? Renaming like `com.ververica.cdc` to `org.apache.flink.cdc` might be necessary. I have rebase the...

[fix-issue-2676] repair a snapshot-split bug:

> Hi @AidenPerce, the Db2 CI fix has been merged into `master` branch, could you please rebase this PR and see if this problem persists? Thank you! I get a...

[fix-issue-2676] repair a snapshot-split bug:

@yuxiqian may help to merge it please ?

flink-connector-mysql-cdc-2.4.2 when a multi-column primary key index starts with a varchar column, a snapshot chunk can potentially return a large number of rows and cause jvm to OOM.

This bug is caused by different charsets and collation rules in the MySQL database. Because we use JAVA to compare the max value of table and the Chunk-End value to...

flink-connector-mysql-cdc-2.4.2 when a multi-column primary key index starts with a varchar column, a snapshot chunk can potentially return a large number of rows and cause jvm to OOM.

When use a primaryKey which is varchar, we can reproduce this problem; Example: + Create a table with a varchar-primaryKey, charset is "utf8mb4", collation is "utf8mb4_general_ci"; + The values of...

flink-connector-mysql-cdc-2.4.2 when a multi-column primary key index starts with a varchar column, a snapshot chunk can potentially return a large number of rows and cause jvm to OOM.

> Hi, @trikker . Mysql CDC supports to set `scan.incremental.snapshot.chunk.key-column` to select a column in the primary key to split chunks. When the type of the partition key is set...