linyu003 commented on issue #4247:
URL:
https://github.com/apache/incubator-seatunnel/issues/4247#issuecomment-1484169191
I am watching on this issue now. I have reproduced the problem by the
config below.
```
env {
execution.parallelism = 1
job.mode = "STREAMING"
execution.checkpoint.interval = 10000
}
source {
FakeSource {
parallelism = 1
result_table_name = "fake"
split.read-interval = 1000
row.num = 2
schema = {
fields {
name = "string"
age = "int"
}
}
}
}
transform {
sql {
sql = "select name,age from fake"
}
}
sink{
kafka {
topic = "test01"
bootstrap.servers = "localhost:9092"
semantics = EXACTLY_ONCE
}
}
```
It seems that the KafkaSinkCommitter doesnot work well if the transaction is
empty (meaning that the trasaction has no record to commit).
A empty transaction is a special case, so I want to just commit the
transaction in KafkaSinkWriter.snapshotState, instead of working hard to make
KafkaSinkCommitter compatitive with it.
I could submit a pr for this in several days.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
