icebergS3Cluster adds deleted rows

[PavelkoSemen]

Company or project name

No response

Describe what's wrong

If you use icebergS3, everything works correctly, as soon as I start using icebergS3Cluster, I see duplicates by keys. On the Iceberg tables side, this row is marked as deleted.

Does it reproduce on the most recent release?

Yes

How to reproduce

Which ClickHouse server version to use:
25.8.4.13
Queries to run that lead to unexpected result:
TRINO:

create table test.test.test_ch_cluster(
   dataflow_dttm timestamp,
   id integer
);

insert into test.test.test_ch_cluster values 
(current_timestamp,1)


update test.test.test_ch_cluster set dataflow_dttm = current_timestamp;

CH:

SELECT * FROM icebergS3Cluster(standard_cluster,  NAMED_COLLECTION , url='https://s3a', filename ='table')

SELECT * FROM icebergS3(NAMED_COLLECTION , url='https://s3a', filename ='table')

Expected behavior

No response

Error message and/or stacktrace

No response

Additional context

No response

[Ted-Jiang]

+1 facing the same issue, icebergHDFSCluster not apply the deletion files, May i raise the pr to support this?

[divanik]

The fix was merged to master, I will try to backport it to 25.10 version in a couple of days