Refactor the config for log dir of FE and BE
TLDR:
- Use env variable `LOG_DIR` to set root log dir
- Remove `sys_log_dir` for FE and BE
Details:
1. FE
1. The root log dir is set by env variable `LOG_DIR` in `fe.conf`
2. The default value of `audit_log_dir` is same as `${LOG_DIR}/`
3. The default value of `spark_launcher_log_dir` is `${LOG_DIR}/spark_launcher_log`
4. The default value of `nereids_trace_log_dir` is `${LOG_DIR}/nereids_trace_log`
5. The origin `sys_log_dir` is deprecated, and default value is `""`.
But for compatibility, if user already set `sys_log_dir` before, Doris will still use it as root log dir.
2. BE
1. The root log dir is set by env variable `LOG_DIR` in `be.conf`
2. Remove `pipeline_tracing_log_dir`, use `${LOG_DIR}` directly.
3. The origin `sys_log_dir` is deprecated, and default value is `""`.
But for compatibility, if user already set `sys_log_dir` before, Doris will still use it as root log dir.
During the renaming of `JdbcCatalog`, I noticed that the `jdbcClient` was being closed,
resulting in exceptions during subsequent queries. This happens because the `removeCatalog`
method is invoked when changing the name, which in turn calls the `onClose` method of the catalog.
Ideally, the client should not be closed when renaming the catalog.
However, to avoid extra checks in the `removeCatalog` method, we can simply execute `onRefresh`
in the `addCatalog` method to address this issue.
Following #25138, unified schema change interface for parquet and orc reader, and can be applied to other format readers as well.
Unified schema change interface for all format readers:
- First, read the data according to the column type of the file into source column;
- Second, convert source column to the destination column with type planned by FE.
The meta of the restore may come from different clusters, so the
original ID in the meta may conflict with the ID of the new cluster. For
example, if a newly allocated ID happens to be the same as an original ID,
the original one may be overwritten when executing `put`.
Issue Number: #31442
- Add hive-writer runtime profiles.
- Change output file names to `${query_id}${uuid}-${index}.${compression}.${format}`. e.g. `"d8735c6fa444a6d-acd392981e510c2b_34fbdcbb-b2e1-4f2c-b68c-a384238954a9-0.snappy.parquet"`. For the same partition writer, when the file size exceeds `hive_sink_max_file_size`, the currently written file will be closed and a new file will be generated, in which ${index} in the new file name will be incremented, while the rest will be the same .
Issue #31442#32726
1. add LocalDfsFileSystem to manipulate local files.
2. add HMSCachedClientTest to analog HMS services.
3. add test for rollback commit.
Issue #31442
Add `TransactionManager` and `Transaction`.
```
public interface Transaction {
void commit() throws UserException;
void rollback();
}
public interface TransactionManager {
long begin();
void commit(long id) throws UserException;
void rollback(long id);
Transaction getTransaction(long id);
}
```
`TransactionManager` is used to manage all external transactions:
The application layer should manage the entire transaction through this `TransactionManager`, like:
```
transactionManager.commit();
transactionManager.rollback();
```
`Transaction` is an interface. You can implement this interface according to the specific content, such as `HMSTransaction` currently implemented, iceberg that may be implemented in the future, etc.
Issue Number: #31442
dependent on #32824
add ddl(create and drop) test
add ctas test
add complex type test
TODO:
bucketed table test
truncate test
add/drop partition test
