thswlsqls opened a new issue, #11279:
URL: https://github.com/apache/paimon/issues/11279

   **Search before asking**
   - [x] I searched in the [issues](https://github.com/apache/paimon/issues) 
and found nothing similar.
   
   **Paimon version**
   master @ 3a8e01f35d
   
   **Compute Engine**
   Flink
   
   **Minimal reproduce step**
   1. Create an append table with `bucket = -1`, `data-evolution.enabled = 
true` and `row-tracking.enabled = true`, then write some data.
   2. Run the documented historical partition compaction for databases 
(`maintenance/dedicated-compaction.mdx`), in `divided` or `combined` batch mode:
      `CALL sys.compact_database(mode => 'divided', partition_idle_time => '1 
d')`
   3. The latest snapshot of the Data Evolution table is a `COMPACT` snapshot, 
although its data was written seconds ago.
   
   `CompactDatabaseAction.buildForUnawareBucketCompaction()` builds `new 
DataEvolutionTableCompact(env, fullName, table).build()` without passing 
`partitionIdleTime`. Both batch modes reach this method.
   
   **What doesn't meet your expectations?**
   Only partitions idle for `partition_idle_time` should be compacted, so the 
table above should stay untouched (latest snapshot still `APPEND`). #11197 
already fixed this for the single-table `compact` action; `compact_database` 
still drops the argument for Data Evolution tables.
   
   **Anything else?**
   Ordinary append and bucketed tables in the same job already honor it.
   
   **Are you willing to submit a PR?**
   - [x] I'm willing to submit a PR!
   


-- 
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]

Reply via email to