You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@seatunnel.apache.org by "hutiefang76 (via GitHub)" <gi...@apache.org> on 2026/07/02 01:46:23 UTC

[PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

hutiefang76 opened a new pull request, #11271:
URL: https://github.com/apache/seatunnel/pull/11271

   ### Purpose of this pull request
   
   Fixes #11260.
   
   When a multi-table CDC job is restored from a savepoint after one table is removed from the current job configuration, the restored incremental split can still carry state for the removed table. That stale state may later be used while registering database history and schema, causing restore startup to fail for a table that is no longer part of the job.
   
   This patch keeps the fix focused on restored incremental-state pruning:
   
   - prune removed tables from `IncrementalSplit.tableIds`
   - prune matching entries from `completedSnapshotSplitInfos`
   - prune removed-table `checkpointTables`
   - prune removed-table `historyTableChanges`
   - skip a restored incremental split if no tables remain after pruning
   
   ### Does this PR introduce _any_ user-facing change?
   
   No. This only fixes restore behavior for CDC jobs whose table list changed before restoring from a checkpoint/savepoint.
   
   ### How was this patch tested?
   
   Added `IncrementalSplitTest` to verify that a removed table is pruned from all restored incremental-state structures.
   
   Tested locally with Java 8:
   
   ```bash
   JAVA_HOME=$(/usr/libexec/java_home -v 1.8) PATH="$JAVA_HOME/bin:$PATH" ./mvnw -o -pl seatunnel-connectors-v2/connector-cdc/connector-cdc-base clean test -Dtest=IncrementalSplitTest,IncrementalSplitStateTest -DskipITs -Drat.skip=true -DfailIfNoTests=false
   ```
   
   Result: 3 tests, 0 failures, 0 errors.
   
   ```bash
   JAVA_HOME=$(/usr/libexec/java_home -v 1.8) PATH="$JAVA_HOME/bin:$PATH" ./mvnw -o -pl seatunnel-connectors-v2/connector-cdc/connector-cdc-base test -Dtest=IncrementalSourceSplitReaderTest,IncrementalSourceStreamFetcherTest -DskipITs -Drat.skip=true -DfailIfNoTests=false
   ```
   
   Result: 5 tests, 0 failures, 0 errors.
   
   Also ran:
   
   ```bash
   git diff --check
   ```
   
   No output.
   
   ### Check list
   
   * [x] If any new Jar binary package adding in your PR, please add License Notice according
     [New License Guide](https://github.com/apache/seatunnel/blob/dev/docs/en/developer/new-license.md)
   * [x] If necessary, please update the documentation to describe the new feature. https://github.com/apache/seatunnel/tree/dev/docs
   * [x] If necessary, please update `incompatible-changes.md` to describe the incompatibility caused by this PR.
   * [x] If you are contributing the connector code, please check that the following files are updated:
     1. Update [plugin-mapping.properties](https://github.com/apache/seatunnel/blob/dev/plugin-mapping.properties) and add new connector information in it
     2. Update the pom file of [seatunnel-dist](https://github.com/apache/seatunnel/blob/dev/seatunnel-dist/pom.xml)
     3. Add ci label in [label-scope-conf](https://github.com/apache/seatunnel/blob/dev/.github/workflows/labeler/label-scope-conf.yml)
     4. Add e2e testcase in [seatunnel-e2e](https://github.com/apache/seatunnel/tree/dev/seatunnel-e2e/seatunnel-connector-v2-e2e/)
     5. Update connector [plugin_config](https://github.com/apache/seatunnel/blob/dev/config/plugin_config)
   
   The checklist items above are not applicable because this patch does not add a new connector, binary package, user-facing feature, documentation change, or incompatible behavior.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5385675957

   Thanks @SEZ9 — this is a fair, well-evidenced catch, and I want to concede it directly rather than defend my own "Ready to merge" conclusion from earlier today.
   
   I re-checked the current head (`79db105ea259`) myself: `pruneRestoredIncrementalSplit()` (`IncrementalSourceReader.java:246-253`) does call `dataSourceDialect.discoverDataCollections(sourceConfig)` unguarded inside `addSplits()`, with no try/catch and no fallback if it throws or returns a partial result. Before this PR, `addSplits()` for a restored incremental split was a pure in-memory operation; this PR adds a live DB round trip directly onto the restore path. If that call throws on a transient DB issue, the recovery attempt itself fails (restart-loop risk on Zeta/Flink). If it returns a partial/empty set (replica lag, a momentary privilege issue), `pruneTables()` will silently drop `completedSnapshotSplitInfos`/`checkpointTables`/`historyTableChanges` for a table that was never actually removed from the job config, and the next checkpoint persists that loss irreversibly. Both failure modes are real and match what you described.
   
   That is a legitimate High-severity blocker my last review missed — I was focused on confirming the `tableWatermarks` exclusion revert and did not re-examine `discoverDataCollections()`'s error-handling contract on this now-hot restore path. I am retracting the "no blockers" framing from my prior review on this specific point; your Issue 1 is the blocking item here. Issues 2, 3, and 5 (the ad-hoc `TableId` reconstruction, discovery-vs-config as the source of truth, and the metadata-presence restore heuristic) are also worth the author's attention as solid non-blocking follow-ups in the same area.
   
   @hutiefang76 — once Issue 1 is addressed (wrapping the discovery call with a safe fallback, or deriving the captured set from the configured table list instead of live discovery, per SEZ9's suggestion), I will do a fresh full pass over the updated head.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-4885777351

   Rechecked the current head again.
   
   My source-side conclusion is still the same: I do not see a new blocker on the latest CDC restore-pruning logic. The remaining problem is the gate state around it:
   
   - base branch: `dev`
   - head SHA: `1d4087ea83275e055af10e335d34e0bcc792a016`
   - compare status: `diverged`
   - ahead_by: `4`
   - behind_by: `75`
   - current failing lane: `kafka-connector-it (8, ubuntu-latest)`
   - CI reference: https://github.com/hutiefang76/seatunnel/actions/runs/28692362211
   
   That failure is outside the CDC path touched by this PR, and the branch is now far behind the latest base, so the lowest-cost next step is to sync the latest `dev` and rerun CI first. If the rebased head still fails, please share the updated failing check link and I can narrow it down on top of the refreshed revision.
   


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "det101 (via GitHub)" <gi...@apache.org>.
det101 commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3811132359


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/reader/IncrementalSourceReader.java:
##########
@@ -146,7 +147,16 @@ public void addSplits(List<SourceSplitBase> splits) {
                     unfinishedSplits.add(split);
                 }
             } else {
-                unfinishedSplits.add(split.asIncrementalSplit());
+                IncrementalSplit incrementalSplit =
+                        pruneRestoredIncrementalSplit(split.asIncrementalSplit());
+                if (incrementalSplit.getTableIds().isEmpty()) {
+                    log.info(
+                            "subtask {} skip restored incremental split {} because all tables have been removed from current configuration.",
+                            subtaskId,
+                            incrementalSplit.splitId());
+                } else {
+                    unfinishedSplits.add(incrementalSplit);
+                }

Review Comment:
   When a restored incremental split is pruned to empty `tableIds`, this branch skips the split. If that leaves `unfinishedSplits` empty, the existing `else` below sets `needSendSplitRequest = true`, so the reader will ask the enumerator for a new split.
   
   On that path, `IncrementalSplitAssigner.getNext()` → `createIncrementalSplits()` builds a `List<TableId>[incrementalParallelism]` and always calls `createIncrementalSplit(capturedTable, …)` for every slot. If remaining captured tables are fewer than `incrementalParallelism` (including zero), some slots stay `null`, and `createIncrementalSplit` NPEs on `capturedTables.contains(...)`.
   
   Also, restore does not populate `assignedSplits` from reader-held incremental splits, so re-creating splits here may re-assign tables that another restored reader is already consuming.
   
   Typical issue #11260 case (`parallelism = 1`, remove some but not all tables) usually keeps a non-empty split and avoids this path; the risk shows up more when incremental parallelism > remaining tables, or when all tables on one reader’s split are removed.
   
   Worth either skipping null partitions in `createIncrementalSplits`, or signaling no-more-splits instead of re-requesting when prune empties the restored split.



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "nielifeng (via GitHub)" <gi...@apache.org>.
nielifeng commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5139942834

   Runtime verification completed for head `4496c6e4886976ca36945642d2f997bcdcb501a9` against current `dev` `7e25769a8e6fa4de5f65f20fa62e0e3dcc32993c`.
   
   Environment and flow:
   
   - 3-master / 3-worker Zeta cluster, 24 total slots
   - MySQL-CDC two-table baseline, 10 rows per table
   - checkpoint interval 5 seconds, parallelism 1
   - savepoint stop, restore with one table removed from the current config
   - post-restore increment written to both the retained and removed source tables
   
   Verification results:
   
   - targeted tests: `IncrementalSplitTest,IncrementalSplitStateTest` -> 3 tests, 0 failures, 0 errors
   - deployed REST metadata: `gitCommitAbbrev=4496c6e`
   - baseline snapshot: source/sink `10/10` for both tables
   - savepoint: `SAVEPOINT_DONE`
   - restored job: `RUNNING`
   - final checkpoints: `39 triggered / 39 completed / 0 failed / 0 in progress`
   - retained table: source/sink `10 -> 11`
   - removed table: source `10 -> 11`, sink stayed `10`
   - Worker runtime split after restore contained only the retained table in all three relevant structures:
     - `tableIds`
     - `checkpointTables`
     - `historyTableChanges`
   - no `Can't obtain schema` or restore ERROR was observed
   
   For comparison, current `dev` already produced a retained-only `tableIds` list in this non-destructive scenario, but still carried both tables in `checkpointTables` and `historyTableChanges`. The PR removed that stale metadata at the actual reader restore path.
   
   Conclusion: no runtime blocker found on this head. The cluster was rolled back to the default `dev` release after the test and verified at 3 workers, 24/24 unassigned slots, and zero running/pending/failed jobs. Current GitHub checks are green.
   


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "hutiefang76 (via GitHub)" <gi...@apache.org>.
hutiefang76 commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5260406966

   Addressed in `1ede24424`.
   
   I added reader-level coverage for both restore branches through `IncrementalSourceReader.addSplits()` and `pollNext()`:
   
   - when some tables remain, the reader queues the pruned split and snapshots only the retained table;
   - when all tables are removed, the reader queues nothing, does not request synchronously from `addSplits()`, and sends the next split request from the following `pollNext()`.
   
   I also added the suggested Javadoc to `IncrementalSplit.pruneTables(...)`.
   
   Verification:
   - counterfactual run against the old `addSplits()` behavior: 2 tests, 2 failures;
   - related CDC reader/split group: 10 tests, 0 failures;
   - full `connector-cdc-base` module: 60 tests, 0 failures; Spotless passed.
   
   Could you please take another look?


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5350295450

   One more note tying this together for anyone landing on this PR: @davidzollo's `CHANGES_REQUESTED` review from 2026-08-11 flagged missing regression coverage for the reader-side restore control flow (`IncrementalSourceReader.java:149-173` — the "prune, then skip-queue-and-request-a-new-split-if-empty" branch), since the tests at that point only covered `IncrementalSplit.pruneTables()` as a pure data helper. That gap has since been closed by `IncrementalSourceReaderTest.java` (added in a later commit) and, in this latest `31c3bf8374` commit, by the new `IncrementalSplitAssignerTest` cases that exercise the enumerator-side split-recreation this reader branch hands off to.
   
   From my side, after today's full re-review I don't see any remaining source-level blocker, and I've submitted an `APPROVED` review reflecting that. Since I don't have write-level review controls on this repository, my approval doesn't automatically clear the historical `CHANGES_REQUESTED` entry from GitHub's review-decision summary — a maintainer with write access (or @davidzollo himself, since the underlying concern is now addressed) would need to dismiss or update that review for the merge gate to reflect the current state. The only other thing standing between this PR and a green `Build` is a one-off Windows Hazelcast startup flake in an unrelated engine-server test (details in my review above) — rerunning that single job should clear it.
   


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "hutiefang76 (via GitHub)" <gi...@apache.org>.
hutiefang76 commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5397492520

   Rebased the full PR onto current `dev` (now including #11954's engine-server timeout-constant fix). The CDC change set is retained, and I reran the focused CDC tests locally under JDK 8: 11 tests passed across `IncrementalSourceReaderTest`, `IncrementalSplitTest`, and `IncrementalSplitAssignerTest`. A fresh CI run has been triggered by this branch update.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-4957129849

   Rechecked the latest head: the source-level blocker from my previous Daniel round is still closed for me, and the current `Build` gate is now green on this unchanged head.
   
   At this point I do not see a new code-side blocker from Daniel on this revision. For queue bookkeeping, this branch is still `diverged` against the latest `dev` (`ahead_by=1`, `behind_by=43`), but because the required `Build` signal is green now, I am not asking for a sync-only retry from Daniel's side here.
   
   If the repository still needs a write-capable maintainer to refresh the formal review state or dismiss an older visible `changes requested` entry, that final gate may need to be handled separately from Daniel's source-level recheck.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5552702127

   `Build` on `151a7bf49` (fork run `hutiefang76/seatunnel` 33963292524, the dev-sync head) completed **green** — every job success/skipped, no failures. Nothing outstanding from my side; this one is ready for a committer's approve/merge.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3812230716


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/reader/IncrementalSourceReader.java:
##########
@@ -146,7 +147,16 @@ public void addSplits(List<SourceSplitBase> splits) {
                     unfinishedSplits.add(split);
                 }
             } else {
-                unfinishedSplits.add(split.asIncrementalSplit());
+                IncrementalSplit incrementalSplit =
+                        pruneRestoredIncrementalSplit(split.asIncrementalSplit());
+                if (incrementalSplit.getTableIds().isEmpty()) {
+                    log.info(
+                            "subtask {} skip restored incremental split {} because all tables have been removed from current configuration.",
+                            subtaskId,
+                            incrementalSplit.splitId());
+                } else {
+                    unfinishedSplits.add(incrementalSplit);
+                }

Review Comment:
   Nice catch — I traced the full restore round-trip (SourceFlowLifeCycle.restoreState -> RestoredSplitOperation -> IncrementalSourceEnumerator.addSplitsBack -> IncrementalSplitAssigner.addSplits/getNext) and can confirm the NPE at IncrementalSplitAssigner.java:273 (current dev) is reachable exactly the way you describe. Full write-up with root-cause trace and severity is in my review below; raised as Issue 1 (High), credited to you.



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "nzw921rx (via GitHub)" <gi...@apache.org>.
nzw921rx commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-4861605115

   @hutiefang76 Thank you for your contribution 🚀 
   please enable CI:https://github.com/apache/seatunnel/pull/11271/checks?check_run_id=84674499375


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "nzw921rx (via GitHub)" <gi...@apache.org>.
nzw921rx commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3891512246


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/split/IncrementalSplit.java:
##########
@@ -131,4 +135,48 @@ public IncrementalSplit(
         this.checkpointTables = checkpointTables;
         this.historyTableChanges = historyTableChanges;
     }
+
+    /** Returns restored checkpoint state limited to the tables captured by the current job. */
+    public IncrementalSplit pruneTables(Collection<TableId> capturedTables) {
+        Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
+        List<TableId> filteredTableIds =
+                tableIds.stream().filter(capturedTableSet::contains).collect(Collectors.toList());
+        List<CompletedSnapshotSplitInfo> filteredCompletedSnapshotSplitInfos =
+                completedSnapshotSplitInfos.stream()
+                        .filter(info -> capturedTableSet.contains(info.getTableId()))
+                        .collect(Collectors.toList());
+        List<CatalogTable> filteredCheckpointTables =
+                checkpointTables == null
+                        ? null
+                        : checkpointTables.stream()
+                                .filter(
+                                        table ->
+                                                capturedTableSet.contains(

Review Comment:
   Thank you for your contribution. 
   
   1. In the context of DB2, it seems that filtering is not possible here.
   
   DB2 tableId:
   
   ```java
   private static Map<TableId, CatalogTable> createDb2TableMap(
           List<CatalogTable> catalogTables) {
       Map<TableId, CatalogTable> tables = new HashMap<>();
       CatalogTableUtils.convertTables(catalogTables)
               .forEach(
                       (tableId, catalogTable) -> {
                           tables.put(tableId, catalogTable);
                           tables.put(toDb2TableId(tableId), catalogTable);
                       });
       return tables;
   }
   
   private static TableId toDb2TableId(TableId tableId) {
       return new TableId("", tableId.schema(), tableId.table());
   }
   ```
   
   Some other case codes processed by DB2:
   
   ```java
       /**
        * Db2 Debezium metadata always uses an empty catalog because one connector instance captures a
        * single configured database. Explicit SeaTunnel table names still carry the database segment,
        * so startup validation needs to drop that catalog part before comparing with capture tables.
        */
       static TableId toConfiguredDb2TableId(String configuredTable) {
           int firstDot = configuredTable.indexOf('.');
           int secondDot = configuredTable.indexOf('.', firstDot + 1);
           if (firstDot < 0 || secondDot < 0 || secondDot == configuredTable.length() - 1) {
               throw new SeaTunnelException(
                       "DB2 CDC table-names must use database.schema.table format, but found: "
                               + configuredTable);
           }
           return new TableId(
                   "",
                   configuredTable.substring(firstDot + 1, secondDot),
                   configuredTable.substring(secondDot + 1));
       }
   ```
   
   
   2. `JdbcDataSourceDialect` I think the logic for generating tableId should use a unified interface here, and different databases can have personalized processing methods
   
   Suggested change:
   
   ```java
   // JdbcDataSourceDialect
   default TableId toTableId(TablePath tablePath) {
       return new TableId(
               tablePath.getDatabaseName(),
               tablePath.getSchemaName(),
               tablePath.getTableName());
   }
   ```
   
   ```java
   // Db2Dialect
   @Override
   public TableId toTableId(TablePath tablePath) {
       return new TableId(
               "",
               tablePath.getSchemaName(),
               tablePath.getTableName());
   }
   ```
   
   ```java
   // IncrementalSplit
   public IncrementalSplit pruneTables(
           Collection<TableId> capturedTables,
           Function<TablePath, TableId> tableIdConverter) {
   
       Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
   
       List<CatalogTable> filteredCheckpointTables =
               checkpointTables == null
                       ? null
                       : checkpointTables.stream()
                               .filter(
                                       table ->
                                               capturedTableSet.contains(
                                                       tableIdConverter.apply(
                                                               table.getTablePath())))
                               .collect(Collectors.toList());
   
       // Keep the remaining filtering logic unchanged.
   }
   ```
   
   ```java
   // IncrementalSourceReader
   IncrementalSplit prunedSplit =
           incrementalSplit.pruneTables(
                   capturedTables,
                   dataSourceDialect::toTableId);
   ```
   



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "nzw921rx (via GitHub)" <gi...@apache.org>.
nzw921rx commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3891512246


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/split/IncrementalSplit.java:
##########
@@ -131,4 +135,48 @@ public IncrementalSplit(
         this.checkpointTables = checkpointTables;
         this.historyTableChanges = historyTableChanges;
     }
+
+    /** Returns restored checkpoint state limited to the tables captured by the current job. */
+    public IncrementalSplit pruneTables(Collection<TableId> capturedTables) {
+        Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
+        List<TableId> filteredTableIds =
+                tableIds.stream().filter(capturedTableSet::contains).collect(Collectors.toList());
+        List<CompletedSnapshotSplitInfo> filteredCompletedSnapshotSplitInfos =
+                completedSnapshotSplitInfos.stream()
+                        .filter(info -> capturedTableSet.contains(info.getTableId()))
+                        .collect(Collectors.toList());
+        List<CatalogTable> filteredCheckpointTables =
+                checkpointTables == null
+                        ? null
+                        : checkpointTables.stream()
+                                .filter(
+                                        table ->
+                                                capturedTableSet.contains(

Review Comment:
   Thank you for your contribution. 
   
   1. In the context of DB2, it seems that filtering is not possible here.
   
   DB2 tableId:
   
   ```java
   private static Map<TableId, CatalogTable> createDb2TableMap(
           List<CatalogTable> catalogTables) {
       Map<TableId, CatalogTable> tables = new HashMap<>();
       CatalogTableUtils.convertTables(catalogTables)
               .forEach(
                       (tableId, catalogTable) -> {
                           tables.put(tableId, catalogTable);
                           tables.put(toDb2TableId(tableId), catalogTable);
                       });
       return tables;
   }
   
   private static TableId toDb2TableId(TableId tableId) {
       return new TableId("", tableId.schema(), tableId.table());
   }
   ```
   
   Some other case codes processed by DB2:
   
   ```java
       /**
        * Db2 Debezium metadata always uses an empty catalog because one connector instance captures a
        * single configured database. Explicit SeaTunnel table names still carry the database segment,
        * so startup validation needs to drop that catalog part before comparing with capture tables.
        */
       static TableId toConfiguredDb2TableId(String configuredTable) {
           int firstDot = configuredTable.indexOf('.');
           int secondDot = configuredTable.indexOf('.', firstDot + 1);
           if (firstDot < 0 || secondDot < 0 || secondDot == configuredTable.length() - 1) {
               throw new SeaTunnelException(
                       "DB2 CDC table-names must use database.schema.table format, but found: "
                               + configuredTable);
           }
           return new TableId(
                   "",
                   configuredTable.substring(firstDot + 1, secondDot),
                   configuredTable.substring(secondDot + 1));
       }
   ```
   
   
   2. `JdbcDataSourceDialect` I think the logic for generating tableId should use a unified interface here, and different databases can have personalized processing methods
   
   exapmles:
   
   ```java
   default TableId toTableId(TablePath tablePath) {
       return new TableId(
               tablePath.getDatabaseName(),
               tablePath.getSchemaName(),
               tablePath.getTableName());
   }
   
   // DB2
   @Override
   public TableId toTableId(TablePath tablePath) {
       return new TableId(
               "",
               tablePath.getSchemaName(),
               tablePath.getTableName());
   }
   ```



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "davidzollo (via GitHub)" <gi...@apache.org>.
davidzollo commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3896019753


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/split/IncrementalSplit.java:
##########
@@ -131,4 +135,48 @@ public IncrementalSplit(
         this.checkpointTables = checkpointTables;
         this.historyTableChanges = historyTableChanges;
     }
+
+    /** Returns restored checkpoint state limited to the tables captured by the current job. */
+    public IncrementalSplit pruneTables(Collection<TableId> capturedTables) {
+        Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
+        List<TableId> filteredTableIds =
+                tableIds.stream().filter(capturedTableSet::contains).collect(Collectors.toList());
+        List<CompletedSnapshotSplitInfo> filteredCompletedSnapshotSplitInfos =
+                completedSnapshotSplitInfos.stream()
+                        .filter(info -> capturedTableSet.contains(info.getTableId()))
+                        .collect(Collectors.toList());
+        List<CatalogTable> filteredCheckpointTables =
+                checkpointTables == null
+                        ? null
+                        : checkpointTables.stream()
+                                .filter(
+                                        table ->
+                                                capturedTableSet.contains(

Review Comment:
   > Thank you for your contribution.
   > 
   > 1. In the context of DB2, it seems that filtering is not possible here.
   > 
   > DB2 tableId:
   > 
   > ```java
   > private static Map<TableId, CatalogTable> createDb2TableMap(
   >         List<CatalogTable> catalogTables) {
   >     Map<TableId, CatalogTable> tables = new HashMap<>();
   >     CatalogTableUtils.convertTables(catalogTables)
   >             .forEach(
   >                     (tableId, catalogTable) -> {
   >                         tables.put(tableId, catalogTable);
   >                         tables.put(toDb2TableId(tableId), catalogTable);
   >                     });
   >     return tables;
   > }
   > 
   > private static TableId toDb2TableId(TableId tableId) {
   >     return new TableId("", tableId.schema(), tableId.table());
   > }
   > ```
   > 
   > Some other case codes processed by DB2:
   > 
   > ```java
   >     /**
   >      * Db2 Debezium metadata always uses an empty catalog because one connector instance captures a
   >      * single configured database. Explicit SeaTunnel table names still carry the database segment,
   >      * so startup validation needs to drop that catalog part before comparing with capture tables.
   >      */
   >     static TableId toConfiguredDb2TableId(String configuredTable) {
   >         int firstDot = configuredTable.indexOf('.');
   >         int secondDot = configuredTable.indexOf('.', firstDot + 1);
   >         if (firstDot < 0 || secondDot < 0 || secondDot == configuredTable.length() - 1) {
   >             throw new SeaTunnelException(
   >                     "DB2 CDC table-names must use database.schema.table format, but found: "
   >                             + configuredTable);
   >         }
   >         return new TableId(
   >                 "",
   >                 configuredTable.substring(firstDot + 1, secondDot),
   >                 configuredTable.substring(secondDot + 1));
   >     }
   > ```
   > 
   > 2. `JdbcDataSourceDialect` I think the logic for generating tableId should use a unified interface here, and different databases can have personalized processing methods
   > 
   > Suggested change:
   > 
   > ```java
   > // JdbcDataSourceDialect
   > default TableId toTableId(TablePath tablePath) {
   >     return new TableId(
   >             tablePath.getDatabaseName(),
   >             tablePath.getSchemaName(),
   >             tablePath.getTableName());
   > }
   > ```
   > 
   > ```java
   > // Db2Dialect
   > @Override
   > public TableId toTableId(TablePath tablePath) {
   >     return new TableId(
   >             "",
   >             tablePath.getSchemaName(),
   >             tablePath.getTableName());
   > }
   > ```
   > 
   > ```java
   > // IncrementalSplit
   > public IncrementalSplit pruneTables(
   >         Collection<TableId> capturedTables,
   >         Function<TablePath, TableId> tableIdConverter) {
   > 
   >     Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
   > 
   >     List<CatalogTable> filteredCheckpointTables =
   >             checkpointTables == null
   >                     ? null
   >                     : checkpointTables.stream()
   >                             .filter(
   >                                     table ->
   >                                             capturedTableSet.contains(
   >                                                     tableIdConverter.apply(
   >                                                             table.getTablePath())))
   >                             .collect(Collectors.toList());
   > 
   >     // Keep the remaining filtering logic unchanged.
   > }
   > ```
   > 
   > ```java
   > // IncrementalSourceReader
   > IncrementalSplit prunedSplit =
   >         incrementalSplit.pruneTables(
   >                 capturedTables,
   >                 dataSourceDialect::toTableId);
   > ```
   
   @hutiefang76 Can you answer these questions from @nzw921rx ? 



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "hutiefang76 (via GitHub)" <gi...@apache.org>.
hutiefang76 commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5371638523

   Addressed the blocking restore regression in 79db105ea.
   
   I removed the two tableWatermarks-based exclusions. A restored table must still be included in the newly assigned incremental split; tableWatermarks supplies that splits resumed startup offset rather than marking the table as completed.
   
   The regression test now restores a split, requests the next split, and verifies both that the table is reassigned and that its restored startup offset is preserved.
   
   Verified locally with JDK 8:
   - ./mvnw -pl seatunnel-connectors-v2/connector-cdc/connector-cdc-base -Dtest=IncrementalSplitAssignerTest test (4 tests, 0 failures/errors)
   - ./mvnw -pl seatunnel-connectors-v2/connector-cdc/connector-cdc-base spotless:check
   - git diff --check
   
   The earlier empty-bucket handling remains unchanged. Thank you for the detailed review; could you please re-review the current head when convenient?


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "hutiefang76 (via GitHub)" <gi...@apache.org>.
hutiefang76 commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5386877953

   Addressed the restore-path safety concern in e381b8ebe.\n\n- Captured-table discovery is attempted at most once per addSplits() batch.\n- If discovery throws, the restored incremental split is kept unchanged, so a transient database failure cannot fail recovery or discard checkpoint metadata.\n- If discovery returns an empty result for a non-empty restored split, pruning is skipped for the same reason.\n- Existing pruning remains in place when discovery returns a non-empty captured-table set.\n\nValidation on JDK 8:\n- ./mvnw -pl seatunnel-connectors-v2/connector-cdc/connector-cdc-base -Dtest=IncrementalSourceReaderTest test (4 tests, 0 failures/errors)\n- ./mvnw -pl seatunnel-connectors-v2/connector-cdc/connector-cdc-base test (73 tests, 0 failures/errors)\n\nThe regression coverage includes discovery failure, empty discovery, and one discovery call for a batch containing multiple restored incremental splits. Please let me know if a fresh review finds any remaining concern.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "davidzollo (via GitHub)" <gi...@apache.org>.
davidzollo commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5509675025

   Thanks for the thorough review. I re-verified each of your 8 findings against the current head (was ee2a1424199, now 4bd75ee8128 after a fix I'm pushing as part of this pass) and am leaving this review open (not dismissing) since I'm addressing part of it myself in this same pass.
   
   **Issue 1 (Blocking) — unguarded live `discoverDataCollections()` call: RESOLVED**
   `IncrementalSourceReader` now has a `discoverCapturedTables()` helper that wraps the dialect call in try/catch and returns `null` on failure (IncrementalSourceReader.java:260-270). `pruneRestoredIncrementalSplit()` returns the split unpruned when `capturedTables == null` (a discovery failure) or when discovery returned empty while the split still has tableIds (IncrementalSourceReader.java:272-283), so a transient DB error or empty/partial discovery result no longer crashes the reader or silently discards checkpoint state. `addSplits()` also only calls this once per batch via the `capturedTablesDiscovered` flag (IncrementalSourceReader.java:133-134, 154-157), addressing the "repeated metadata queries" note too. Covered by `IncrementalSourceReaderTest#testAddSplitsKeepsRestoredSplitWhenDiscoveryFails`, `#testAddSplitsKeepsRestoredSplitWhenDiscoveryReturnsEmpty`, and `#testAddSplitsDiscoversCapturedTablesOnlyOncePerBatch`.
   
   **Issue 2 (Medium) — ad-hoc `TableId` reconstruction fragile across dialects: RESOLVED**
   Fixed in commit 7df41cf5d9021a9d00ec05e4f7dda8321ebd80ea (already on head before this pass): `DataSourceDialect.toTableId(TablePath)` default method + `Db2Dialect` override for the empty-catalog form, threaded through `IncrementalSplit.pruneTables(capturedTables, tableIdConverter)` and called as `dataSourceDialect::toTableId` from the reader. Covered by `IncrementalSplitTest#testPruneTablesUsesDialectSpecificTableIdConverter` and `Db2IncrementalSourceFactoryTest`. (This is the same issue nzw921rx raised concretely for Db2; I dismissed that review with the same evidence.)
   
   **Issue 3 (Medium) — pruning keyed off live discovery, INFO-level destructive logging: PARTIALLY RESOLVED**
   The destructive part is fixed: a transient empty/failed discovery no longer prunes anything (see Issue 1). Two of your suggestions are still open and I'm not taking them on in this pass: (a) the empty-split skip log at IncrementalSourceReader.java:161-165 and the prune-summary log at 287-292 are still `log.info`, not `log.warn`; (b) pruning still keys off `discoverDataCollections()` rather than the configured table filter. Leaving both as non-blocking follow-ups per your own triage.
   
   **Issue 4 (Medium) — missing incompatible-changes.md entry: RESOLVED**
   Added to both `docs/en/introduction/concepts/incompatible-changes.md` and `docs/zh/introduction/concepts/incompatible-changes.md`, describing the restore-time pruning behavior and the discovery-failure fallback.
   
   **Issue 5 (Medium) — `hasRestoredCheckpointMetadata` heuristic gate: OPEN, non-blocking**
   Unchanged — still infers "restored" from `checkpointDataType != null || checkpointTables non-empty || historyTableChanges non-empty` (IncrementalSourceReader.java:296-302), rather than an explicit restored flag. I traced `snapshotCheckpointDataType()` (IncrementalSourceReader.java:335-350): `checkpointTables` is set from `debeziumDeserializationSchema.getProducedType()` on every checkpoint while in incremental phase, so in practice this should be non-empty once at least one record has been produced for the split's tables. Whether it can be empty on an early checkpoint before any change event has flowed (leaving pruning skipped for that specific restore) is the kind of restore-timing edge case I don't want to redesign without being able to run this locally against a real checkpoint/restore cycle — recommend keeping this open as a dedicated follow-up rather than guessing at a fix here.
   
   **Issue 6 (Medium) — `pruneTables()` unguarded `tableIds`/`completedSnapshotSplitInfos`: FIXED IN THIS PASS**
   Just pushed commit 4bd75ee8128903c57c647b5259bb6725e1509c6a: both are now null-guarded the same way `checkpointTables`/`historyTableChanges` already were, falling back to an empty list instead of NPEing. Added `IncrementalSplitTest#testPruneTablesToleratesNullTableIdsAndCompletedSnapshotSplitInfos` to cover it.
   
   **Issue 7 (Low) — thin Javadoc: RESOLVED**
   `pruneTables()` now has a full Javadoc with `@param`/`@return` and an explanation of the `TablePath`/`TableId` namespace mismatch this method bridges (IncrementalSplit.java:141-151).
   
   **Issue 8 (Low) — log message casing/verbosity: OPEN, non-blocking**
   Unchanged — the skip message still starts lowercase ("subtask {} skip ...", IncrementalSourceReader.java:162-165) and the prune-summary log still dumps full before/after tableId lists (IncrementalSourceReader.java:287-292). Leaving as a non-blocking style nit per your own triage.
   
   Summary: your sole Blocking item (Issue 1) and the two other substantive correctness issues (2, 6) are resolved with the evidence above; docs (4) and Javadoc (7) are done. Issues 3 and 8 have partial/no action (logging nits, explicitly non-blocking in your own conclusion). Issue 5 is a real open question I'm flagging rather than guessing at, since I have no way to verify a redesign of the restore-detection gate without running this against a live checkpoint/restore cycle. Given your own conclusion was "can merge after the blocking items are fixed" and Issue 1 is now resolved, I'd appreciate you taking another pass at the current head (4bd75ee8128) when you have a chance — I'm leaving this review open rather than dismissing it myself since I just pushed a fix touching part of it in this same pass.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5409164774

   Thanks for the quick turnaround, @hutiefang76 — and nicely done tracking that down.
   
   I checked the current head (`96141c4267`) directly: it does include `43fe63b1fc` (#11954, the `FAILED_JOB_EVENT_TIMEOUT_SECONDS` fix) in its history, so the rebase landed exactly where I suggested in my last round. My approval from earlier today was already on this exact commit, so nothing changes there on my end — the source-level review still stands (byte-identical CDC logic to what I traced line-by-line, docs entries closing Issue 4).
   
   On the local test run: `IncrementalSourceReaderTest`, `IncrementalSplitTest`, and `IncrementalSplitAssignerTest` are exactly the three test classes covering the changed `connector-cdc-base` logic, so 11/11 passing locally lines up with what I'd expect given the source is unchanged from the version I already approved.
   
   One thing worth flagging while it's fresh: I looked at the fork's `Build` run for this head (`hutiefang76/seatunnel` run `32745169272`) and it's still landing on `cancelled` rather than a clean `success`, across a couple of attempts. Everything that did complete came back green — including `unit-test (8, ubuntu-latest)`, which is where these three CDC test classes would run — but `mysql-cdc-connector-it`, `paimon-connector-it`, `elasticsearch-connector-it`, `unit-test (11, ubuntu-latest)`, `unit-test (11, windows-latest)`, and `Build website` all got cut off mid-run rather than finishing (attempt 2 cancelled `mysql-cdc-connector-it`/`doris-connector-it`/`paimon-connector-it (11)`, attempt 3's re-run of the survivors cancelled again). Nothing in what I can see points to an actual test failure — it looks like the run is getting cut off rather than red — but it's worth one more clean rerun (or splitting the retry across a bit more time) so the `Build` check actually reflects 
 a finished state before merge, since a `cancelled` conclusion won't satisfy the required check either way.
   
   To be clear on where things stand: my side has no outstanding blockers on the code itself. The one open item on the PR is @SEZ9's CHANGES_REQUESTED review (Issue 1 there was resolved back in `e381b8eb`, and the remaining items 2/3/5/6/7/8 in my summary table are non-blocking, in my view) — I don't have dismiss rights on someone else's review, so that one will need their own follow-up or a maintainer call. Really appreciate you sticking with this through this many rounds.
   


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3891697885


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/split/IncrementalSplit.java:
##########
@@ -131,4 +135,48 @@ public IncrementalSplit(
         this.checkpointTables = checkpointTables;
         this.historyTableChanges = historyTableChanges;
     }
+
+    /** Returns restored checkpoint state limited to the tables captured by the current job. */
+    public IncrementalSplit pruneTables(Collection<TableId> capturedTables) {
+        Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
+        List<TableId> filteredTableIds =
+                tableIds.stream().filter(capturedTableSet::contains).collect(Collectors.toList());
+        List<CompletedSnapshotSplitInfo> filteredCompletedSnapshotSplitInfos =
+                completedSnapshotSplitInfos.stream()
+                        .filter(info -> capturedTableSet.contains(info.getTableId()))
+                        .collect(Collectors.toList());
+        List<CatalogTable> filteredCheckpointTables =
+                checkpointTables == null
+                        ? null
+                        : checkpointTables.stream()
+                                .filter(
+                                        table ->
+                                                capturedTableSet.contains(

Review Comment:
   Thanks for the detailed catch and the concrete DB2 code — I checked this against the current head (`83a904f4c0`) and your concern is valid, and I'd actually raise the severity above where I had it tracked.
   
   Confirmed root cause: `IncrementalSplit.pruneTables()` (`IncrementalSplit.java:148-162`) rebuilds a Debezium `TableId` straight from `CatalogTable#getTablePath()`:
   
   ```java
   new TableId(
       table.getTablePath().getDatabaseName(),
       table.getTablePath().getSchemaName(),
       table.getTablePath().getTableName())
   ```
   
   For Db2 this always mismatches `capturedTables`, because `Db2Dialect.discoverDataCollections()` returns tables straight from `listOfChangeTables()`/`TableDiscoveryUtils.listTables()`, and both `Db2Dialect.toDb2TableId()` and `TableDiscoveryUtils.toConfiguredDb2TableId()` document exactly why:
   
   > "Db2 Debezium metadata always uses an empty catalog because one connector instance captures a single configured database. SeaTunnel catalog tables keep the database name, so all runtime lookups are normalized before comparing with Debezium metadata."
   
   So `capturedTableSet` for Db2 always carries `TableId("", schema, table)`, while the ad-hoc reconstruction in `pruneTables()` produces `TableId(realDbName, schema, table)`. `TableId.equals()` includes the catalog, so `capturedTableSet.contains(...)` is false for every entry — not just for removed tables. This is worse than the "can mismatch" framing I had for this as a non-blocking Medium item in my earlier reviews: for Db2 it isn't a partial/edge case, it deterministically wipes `checkpointTables` to an empty (but non-null) list on every single incremental-phase restore that carries checkpoint tables, since `snapshotCheckpointDataType()` populates `checkpointTables` unconditionally on every checkpoint (`IncrementalSourceReader.java:334-336`).
   
   That empty-but-non-null list then propagates further than just internal bookkeeping: `BaseChangeStreamTableSourceFactory.getRestoreTableStruct()` (`BaseChangeStreamTableSourceFactory.java:86-90`) checks `incrementalSplit.getCheckpointTables() != null` (not `isEmpty()`), so it accepts the emptied list as the authoritative restored table struct and returns it to `restoreSource(...)`. For Db2 CDC jobs that rely on this path, that means job restart can lose the checkpoint-carried table struct entirely, not just fail to prune a removed table correctly.
   
   I agree with your fix direction: `pruneTables()` shouldn't reconstruct a generic `TableId` itself. Threading a dialect-aware converter through (your `JdbcDataSourceDialect#toTableId(TablePath)` default + `Db2Dialect` override, passed into `pruneTables`/used from `IncrementalSourceReader`) is the right shape, since it reuses the same normalization Db2Dialect already applies everywhere else (`checkAllTablesEnabledCapture`, `queryTableSchema`, `getPrimaryKey`, `getConstraintKeys`).
   
   @hutiefang76 given the above, I'd treat this as a High-severity blocker for Db2 specifically (not the pre-existing non-blocking Medium item), since it affects every restore rather than only the removed-table case. Worth fixing before merge — thanks again @nzw921rx for pinning down the concrete failure mode with real Db2 code.



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "nzw921rx (via GitHub)" <gi...@apache.org>.
nzw921rx commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3891512246


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/split/IncrementalSplit.java:
##########
@@ -131,4 +135,48 @@ public IncrementalSplit(
         this.checkpointTables = checkpointTables;
         this.historyTableChanges = historyTableChanges;
     }
+
+    /** Returns restored checkpoint state limited to the tables captured by the current job. */
+    public IncrementalSplit pruneTables(Collection<TableId> capturedTables) {
+        Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
+        List<TableId> filteredTableIds =
+                tableIds.stream().filter(capturedTableSet::contains).collect(Collectors.toList());
+        List<CompletedSnapshotSplitInfo> filteredCompletedSnapshotSplitInfos =
+                completedSnapshotSplitInfos.stream()
+                        .filter(info -> capturedTableSet.contains(info.getTableId()))
+                        .collect(Collectors.toList());
+        List<CatalogTable> filteredCheckpointTables =
+                checkpointTables == null
+                        ? null
+                        : checkpointTables.stream()
+                                .filter(
+                                        table ->
+                                                capturedTableSet.contains(

Review Comment:
   Thank you for your contribution. 
   
   1. In the context of DB2, it seems that filtering is not possible here.
   
   DB2 tableId:
   
   ```java
   private static Map<TableId, CatalogTable> createDb2TableMap(
           List<CatalogTable> catalogTables) {
       Map<TableId, CatalogTable> tables = new HashMap<>();
       CatalogTableUtils.convertTables(catalogTables)
               .forEach(
                       (tableId, catalogTable) -> {
                           tables.put(tableId, catalogTable);
                           tables.put(toDb2TableId(tableId), catalogTable);
                       });
       return tables;
   }
   
   private static TableId toDb2TableId(TableId tableId) {
       return new TableId("", tableId.schema(), tableId.table());
   }
   ```
   
   Some other case codes processed by DB2:
   
   ```java
       /**
        * Db2 Debezium metadata always uses an empty catalog because one connector instance captures a
        * single configured database. Explicit SeaTunnel table names still carry the database segment,
        * so startup validation needs to drop that catalog part before comparing with capture tables.
        */
       static TableId toConfiguredDb2TableId(String configuredTable) {
           int firstDot = configuredTable.indexOf('.');
           int secondDot = configuredTable.indexOf('.', firstDot + 1);
           if (firstDot < 0 || secondDot < 0 || secondDot == configuredTable.length() - 1) {
               throw new SeaTunnelException(
                       "DB2 CDC table-names must use database.schema.table format, but found: "
                               + configuredTable);
           }
           return new TableId(
                   "",
                   configuredTable.substring(firstDot + 1, secondDot),
                   configuredTable.substring(secondDot + 1));
       }
   ```
   
   
   2. `JdbcDataSourceDialect` I think the logic for generating tableId should use a unified interface here, and different databases can have personalized processing methods
   
   exapmles:
   
   ```java
   default TableId createTableId(TablePath tablePath) {
       return new TableId(
               tablePath.getDatabaseName(),
               tablePath.getSchemaName(),
               tablePath.getTableName());
   }
   
   // DB2
   @Override
   public TableId toTableId(TablePath tablePath) {
       return new TableId(
               "",
               tablePath.getSchemaName(),
               tablePath.getTableName());
   }
   ```



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "hutiefang76 (via GitHub)" <gi...@apache.org>.
hutiefang76 commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5369791785

   Thanks for flagging this. The latest branch already skips empty round-robin buckets; I added a follow-up at 801c3d9628ebe9c134b513f63cb63d62d54b4bc1 for the other half of the restore path.
   
   When restored incremental splits are added back, their retained table ids are recorded in tableWatermarks. The assigner now removes those ids when computing both the remaining-table set and newly created incremental splits. Therefore, when a reader prunes its restored split to empty and requests another split, the enumerator does not recreate tables that are already being consumed from restored state. getRemainingTables also now reports completion correctly when every captured table is accounted for by restored state.
   
   Added shouldNotReassignTablesRestoredFromCheckpoint, alongside the existing empty-bucket coverage. Verified with JDK 8:
   ./mvnw -pl seatunnel-connectors-v2/connector-cdc/connector-cdc-base -Dtest=IncrementalSplitAssignerTest test
   Result: 4 tests, 0 failures/errors.
   Also passed:
   ./mvnw -pl seatunnel-connectors-v2/connector-cdc/connector-cdc-base spotless:check
   
   Could you please take another look at the unresolved thread?


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "hutiefang76 (via GitHub)" <gi...@apache.org>.
hutiefang76 commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5394820370

   Rebased the branch onto current  () and resolved the documentation-only conflict by retaining the upstream ORC and Google Bigtable entries alongside the CDC restore-state entry.\n\nLocal verification on JDK 8:\n- [INFO] Scanning for projects...
   [INFO] ------------------------------------------------------------------------
   [INFO] Reactor Build Order:
   [INFO] 
   [INFO] SeaTunnel :                                                        [pom]
   [INFO] SeaTunnel : Config :                                               [pom]
   [INFO] SeaTunnel : Config : Base                                          [jar]
   [INFO] SeaTunnel : Config : Shade                                         [jar]
   [INFO] SeaTunnel : Common                                                 [jar]
   [INFO] SeaTunnel : Api                                                    [jar]
   [INFO] SeaTunnel : Connectors V2 :                                        [pom]
   [INFO] SeaTunnel : Connectors V2 : Common                                 [jar]
   [INFO] SeaTunnel : Formats :                                              [pom]
   [INFO] SeaTunnel : Formats : Compatible Debezium Json                     [jar]
   [INFO] SeaTunnel : Connectors V2 : CDC :                                  [pom]
   [INFO] SeaTunnel : Connectors V2 : CDC : Base                             [jar]
   [INFO] 
   [INFO] -------------------< org.apache.seatunnel:seatunnel >-------------------
   [INFO] Building SeaTunnel : 3.0.0-SNAPSHOT                               [1/12]
   [INFO] --------------------------------[ pom ]---------------------------------
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-maven-version) @ seatunnel ---
   [INFO] Rule 0: org.apache.maven.enforcer.rules.version.RequireMavenVersion passed
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-java-version) @ seatunnel ---
   [INFO] Rule 0: org.apache.maven.enforcer.rules.version.RequireJavaVersion passed
   [INFO] 
   [INFO] --- spotless-maven-plugin:2.29.0:check (spotless-check) @ seatunnel ---
   [INFO] Up-to-date checking enabled
   [INFO] Fingerprint mismatch in the index file. Fallback to an empty index
   [INFO] Sorting file /var/folders/kp/rjp70w_s7qz1lwr6044cq8qh0000gn/T/pom3326225135097216879.xml
   [INFO] Pom file is already sorted, exiting
   [INFO] 
   [INFO] --- maven-remote-resources-plugin:3.1.0:process (process-resource-bundles) @ seatunnel ---
   [INFO] Preparing remote bundle org.apache.apache.resources:apache-jar-resource-bundle:1.5
   [INFO] Copying 3 resources from 1 bundle.
   [INFO] 
   [INFO] --- flatten-maven-plugin:1.3.0:flatten (flatten) @ seatunnel ---
   [INFO] Generating flattened POM of project org.apache.seatunnel:seatunnel:pom:3.0.0-SNAPSHOT...
   [INFO] 
   [INFO] --- maven-site-plugin:3.12.1:attach-descriptor (attach-descriptor) @ seatunnel ---
   [INFO] No site descriptor found: nothing to attach.
   [INFO] 
   [INFO] --- maven-failsafe-plugin:2.22.2:integration-test (default) @ seatunnel ---
   [INFO] Tests are skipped.
   [INFO] 
   [INFO] --- maven-failsafe-plugin:2.22.2:verify (default) @ seatunnel ---
   [INFO] Tests are skipped.
   [INFO] 
   [INFO] --- maven-install-plugin:3.1.1:install (default-install) @ seatunnel ---
   [INFO] Installing /private/tmp/seatunnel-11271-fix/.flattened-pom.xml to /Users/hutiefang/.m2/repository/org/apache/seatunnel/seatunnel/3.0.0-SNAPSHOT/seatunnel-3.0.0-SNAPSHOT.pom
   [INFO] 
   [INFO] ---------------< org.apache.seatunnel:seatunnel-config >----------------
   [INFO] Building SeaTunnel : Config : 3.0.0-SNAPSHOT                      [2/12]
   [INFO] --------------------------------[ pom ]---------------------------------
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-maven-version) @ seatunnel-config ---
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-java-version) @ seatunnel-config ---
   [INFO] 
   [INFO] --- spotless-maven-plugin:2.29.0:check (spotless-check) @ seatunnel-config ---
   [INFO] Up-to-date checking enabled
   [INFO] Fingerprint mismatch in the index file. Fallback to an empty index
   [INFO] Sorting file /var/folders/kp/rjp70w_s7qz1lwr6044cq8qh0000gn/T/pom4033479411560198414.xml
   [INFO] Pom file is already sorted, exiting
   [INFO] 
   [INFO] --- maven-remote-resources-plugin:3.1.0:process (process-resource-bundles) @ seatunnel-config ---
   [INFO] Preparing remote bundle org.apache.apache.resources:apache-jar-resource-bundle:1.5
   [INFO] Copying 3 resources from 1 bundle.
   [INFO] 
   [INFO] --- flatten-maven-plugin:1.3.0:flatten (flatten) @ seatunnel-config ---
   [INFO] Generating flattened POM of project org.apache.seatunnel:seatunnel-config:pom:3.0.0-SNAPSHOT...
   [INFO] 
   [INFO] --- maven-site-plugin:3.12.1:attach-descriptor (attach-descriptor) @ seatunnel-config ---
   [INFO] No site descriptor found: nothing to attach.
   [INFO] 
   [INFO] --- maven-failsafe-plugin:2.22.2:integration-test (default) @ seatunnel-config ---
   [INFO] Tests are skipped.
   [INFO] 
   [INFO] --- maven-failsafe-plugin:2.22.2:verify (default) @ seatunnel-config ---
   [INFO] Tests are skipped.
   [INFO] 
   [INFO] --- maven-install-plugin:3.1.1:install (default-install) @ seatunnel-config ---
   [INFO] Installing /private/tmp/seatunnel-11271-fix/seatunnel-config/.flattened-pom.xml to /Users/hutiefang/.m2/repository/org/apache/seatunnel/seatunnel-config/3.0.0-SNAPSHOT/seatunnel-config-3.0.0-SNAPSHOT.pom
   [INFO] 
   [INFO] -------------< org.apache.seatunnel:seatunnel-config-base >-------------
   [INFO] Building SeaTunnel : Config : Base 3.0.0-SNAPSHOT                 [3/12]
   [INFO] --------------------------------[ jar ]---------------------------------
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-maven-version) @ seatunnel-config-base ---
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-java-version) @ seatunnel-config-base ---
   [INFO] 
   [INFO] --- spotless-maven-plugin:2.29.0:check (spotless-check) @ seatunnel-config-base ---
   [INFO] Up-to-date checking enabled
   [INFO] Fingerprint mismatch in the index file. Fallback to an empty index
   [INFO] Sorting file /var/folders/kp/rjp70w_s7qz1lwr6044cq8qh0000gn/T/pom5377268234849462698.xml
   [INFO] Pom file is already sorted, exiting
   [INFO] 
   [INFO] --- maven-remote-resources-plugin:3.1.0:process (process-resource-bundles) @ seatunnel-config-base ---
   [INFO] Preparing remote bundle org.apache.apache.resources:apache-jar-resource-bundle:1.5
   [INFO] Copying 3 resources from 1 bundle.
   [INFO] 
   [INFO] --- maven-resources-plugin:3.3.1:resources (default-resources) @ seatunnel-config-base ---
   [INFO] skip non existing resourceDirectory /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/src/main/resources
   [INFO] Copying 3 resources from target/maven-shared-archive-resources to target/classes
   [INFO] 
   [INFO] --- flatten-maven-plugin:1.3.0:flatten (flatten) @ seatunnel-config-base ---
   [INFO] Generating flattened POM of project org.apache.seatunnel:seatunnel-config-base:jar:3.0.0-SNAPSHOT...
   [INFO] 
   [INFO] --- maven-compiler-plugin:3.10.1:compile (default-compile) @ seatunnel-config-base ---
   [INFO] No sources to compile
   [INFO] 
   [INFO] --- maven-resources-plugin:3.3.1:testResources (default-testResources) @ seatunnel-config-base ---
   [INFO] skip non existing resourceDirectory /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/src/test/resources
   [INFO] Copying 3 resources from target/maven-shared-archive-resources to target/test-classes
   [INFO] 
   [INFO] --- maven-compiler-plugin:3.10.1:testCompile (default-testCompile) @ seatunnel-config-base ---
   [INFO] No sources to compile
   [INFO] 
   [INFO] --- maven-surefire-plugin:2.22.2:test (default-test) @ seatunnel-config-base ---
   [INFO] Tests are skipped.
   [INFO] 
   [INFO] --- maven-jar-plugin:3.3.0:jar (default-jar) @ seatunnel-config-base ---
   [INFO] Building jar: /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/target/seatunnel-config-base-3.0.0-SNAPSHOT.jar
   [INFO] 
   [INFO] --- maven-site-plugin:3.12.1:attach-descriptor (attach-descriptor) @ seatunnel-config-base ---
   [INFO] Skipping because packaging 'jar' is not pom.
   [INFO] 
   [INFO] --- maven-shade-plugin:3.5.1:shade (default) @ seatunnel-config-base ---
   [INFO] Including com.typesafe:config:jar:1.3.3 in the shaded jar.
   [INFO] Including io.prometheus:simpleclient:jar:0.16.0 in the shaded jar.
   [INFO] Including io.prometheus:simpleclient_tracer_otel:jar:0.16.0 in the shaded jar.
   [INFO] Including io.prometheus:simpleclient_tracer_common:jar:0.16.0 in the shaded jar.
   [INFO] Including io.prometheus:simpleclient_tracer_otel_agent:jar:0.16.0 in the shaded jar.
   [INFO] Including io.prometheus:simpleclient_hotspot:jar:0.16.0 in the shaded jar.
   [INFO] Including io.prometheus:simpleclient_httpserver:jar:0.16.0 in the shaded jar.
   [INFO] Including io.prometheus:simpleclient_common:jar:0.16.0 in the shaded jar.
   [INFO] Excluding org.slf4j:slf4j-api:jar:1.7.36 from the shaded jar.
   [INFO] Excluding org.apache.logging.log4j:log4j-slf4j-impl:jar:2.17.1 from the shaded jar.
   [INFO] Excluding org.apache.logging.log4j:log4j-api:jar:2.17.1 from the shaded jar.
   [INFO] Excluding org.apache.logging.log4j:log4j-core:jar:2.17.1 from the shaded jar.
   [INFO] Excluding org.slf4j:jcl-over-slf4j:jar:1.7.36 from the shaded jar.
   [INFO] Excluding org.apache.logging.log4j:log4j-1.2-api:jar:2.17.1 from the shaded jar.
   [INFO] Minimizing jar org.apache.seatunnel:seatunnel-config-base:jar:3.0.0-SNAPSHOT
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT.jar, simpleclient-0.16.0.jar, simpleclient_common-0.16.0.jar, simpleclient_hotspot-0.16.0.jar, simpleclient_httpserver-0.16.0.jar, simpleclient_tracer_common-0.16.0.jar, simpleclient_tracer_otel-0.16.0.jar, simpleclient_tracer_otel_agent-0.16.0.jar define 1 overlapping resource: 
   [WARNING]   - META-INF/MANIFEST.MF
   [WARNING] maven-shade-plugin has detected that some files are
   [WARNING] present in two or more JARs. When this happens, only one
   [WARNING] single version of the file is copied to the uber jar.
   [WARNING] Usually this is not harmful and you can skip these warnings,
   [WARNING] otherwise try to manually exclude artifacts based on
   [WARNING] mvn dependency:tree -Ddetail=true and the above output.
   [WARNING] See https://maven.apache.org/plugins/maven-shade-plugin/
   [INFO] Minimized 265 -> 175 (66%)
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar, simpleclient_tracer_otel-0.16.0-sources.jar define 3 overlapping resources: 
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_tracer_otel/pom.properties
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_tracer_otel/pom.xml
   [WARNING]   - io/prometheus/client/exemplars/tracer/otel/OpenTelemetrySpanContextSupplier.java
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar, simpleclient-0.16.0-sources.jar define 30 overlapping resources: 
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient/pom.properties
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient/pom.xml
   [WARNING]   - io/prometheus/client/CKMSQuantiles.java
   [WARNING]   - io/prometheus/client/Collector.java
   [WARNING]   - io/prometheus/client/CollectorRegistry.java
   [WARNING]   - io/prometheus/client/Counter.java
   [WARNING]   - io/prometheus/client/CounterMetricFamily.java
   [WARNING]   - io/prometheus/client/DoubleAdder.java
   [WARNING]   - io/prometheus/client/Enumeration.java
   [WARNING]   - io/prometheus/client/Environment.java
   [WARNING]   - 20 more...
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar, simpleclient_common-0.16.0-sources.jar define 3 overlapping resources: 
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_common/pom.properties
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_common/pom.xml
   [WARNING]   - io/prometheus/client/exporter/common/TextFormat.java
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar, simpleclient_httpserver-0.16.0-sources.jar define 4 overlapping resources: 
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_httpserver/pom.properties
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_httpserver/pom.xml
   [WARNING]   - io/prometheus/client/exporter/HTTPServer.java
   [WARNING]   - io/prometheus/client/exporter/SampleNameFilterSupplier.java
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar, simpleclient_tracer_common-0.16.0-sources.jar define 3 overlapping resources: 
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_tracer_common/pom.properties
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_tracer_common/pom.xml
   [WARNING]   - io/prometheus/client/exemplars/tracer/common/SpanContextSupplier.java
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar, simpleclient-0.16.0-sources.jar, simpleclient_common-0.16.0-sources.jar, simpleclient_hotspot-0.16.0-sources.jar, simpleclient_httpserver-0.16.0-sources.jar, simpleclient_tracer_common-0.16.0-sources.jar, simpleclient_tracer_otel-0.16.0-sources.jar, simpleclient_tracer_otel_agent-0.16.0-sources.jar define 1 overlapping resource: 
   [WARNING]   - META-INF/MANIFEST.MF
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar, simpleclient_tracer_otel_agent-0.16.0-sources.jar define 3 overlapping resources: 
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_tracer_otel_agent/pom.properties
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_tracer_otel_agent/pom.xml
   [WARNING]   - io/prometheus/client/exemplars/tracer/otel_agent/OpenTelemetryAgentSpanContextSupplier.java
   [WARNING] seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar, simpleclient_hotspot-0.16.0-sources.jar define 11 overlapping resources: 
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_hotspot/pom.properties
   [WARNING]   - META-INF/maven/io.prometheus/simpleclient_hotspot/pom.xml
   [WARNING]   - io/prometheus/client/hotspot/BufferPoolsExports.java
   [WARNING]   - io/prometheus/client/hotspot/ClassLoadingExports.java
   [WARNING]   - io/prometheus/client/hotspot/DefaultExports.java
   [WARNING]   - io/prometheus/client/hotspot/GarbageCollectorExports.java
   [WARNING]   - io/prometheus/client/hotspot/MemoryAllocationExports.java
   [WARNING]   - io/prometheus/client/hotspot/MemoryPoolsExports.java
   [WARNING]   - io/prometheus/client/hotspot/StandardExports.java
   [WARNING]   - io/prometheus/client/hotspot/ThreadExports.java
   [WARNING]   - 1 more...
   [WARNING] maven-shade-plugin has detected that some files are
   [WARNING] present in two or more JARs. When this happens, only one
   [WARNING] single version of the file is copied to the uber jar.
   [WARNING] Usually this is not harmful and you can skip these warnings,
   [WARNING] otherwise try to manually exclude artifacts based on
   [WARNING] mvn dependency:tree -Ddetail=true and the above output.
   [WARNING] See https://maven.apache.org/plugins/maven-shade-plugin/
   [INFO] Minimized 578 -> 488 (84%)
   [INFO] Replacing original artifact with shaded artifact.
   [INFO] Replacing /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/target/seatunnel-config-base-3.0.0-SNAPSHOT.jar with /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/target/seatunnel-config-base-3.0.0-SNAPSHOT-shaded.jar
   [INFO] Replacing original source artifact with shaded source artifact.
   [INFO] Replacing /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/target/seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar with /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/target/seatunnel-config-base-3.0.0-SNAPSHOT-shaded-sources.jar
   [INFO] 
   [INFO] --- build-helper-maven-plugin:3.2.0:attach-artifact (compile) @ seatunnel-config-base ---
   [INFO] 
   [INFO] --- maven-failsafe-plugin:2.22.2:integration-test (default) @ seatunnel-config-base ---
   [INFO] Tests are skipped.
   [INFO] 
   [INFO] --- maven-failsafe-plugin:2.22.2:verify (default) @ seatunnel-config-base ---
   [INFO] Tests are skipped.
   [INFO] 
   [INFO] --- maven-install-plugin:3.1.1:install (default-install) @ seatunnel-config-base ---
   [INFO] Installing /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/.flattened-pom.xml to /Users/hutiefang/.m2/repository/org/apache/seatunnel/seatunnel-config-base/3.0.0-SNAPSHOT/seatunnel-config-base-3.0.0-SNAPSHOT.pom
   [INFO] Installing /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/target/seatunnel-config-base-3.0.0-SNAPSHOT.jar to /Users/hutiefang/.m2/repository/org/apache/seatunnel/seatunnel-config-base/3.0.0-SNAPSHOT/seatunnel-config-base-3.0.0-SNAPSHOT.jar
   [INFO] Installing /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/target/seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar to /Users/hutiefang/.m2/repository/org/apache/seatunnel/seatunnel-config-base/3.0.0-SNAPSHOT/seatunnel-config-base-3.0.0-SNAPSHOT-sources.jar
   [INFO] Installing /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-base/target/seatunnel-config-base-3.0.0-SNAPSHOT.jar to /Users/hutiefang/.m2/repository/org/apache/seatunnel/seatunnel-config-base/3.0.0-SNAPSHOT/seatunnel-config-base-3.0.0-SNAPSHOT-optional.jar
   [INFO] 
   [INFO] ------------< org.apache.seatunnel:seatunnel-config-shade >-------------
   [INFO] Building SeaTunnel : Config : Shade 3.0.0-SNAPSHOT                [4/12]
   [INFO] --------------------------------[ jar ]---------------------------------
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-maven-version) @ seatunnel-config-shade ---
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-java-version) @ seatunnel-config-shade ---
   [INFO] 
   [INFO] --- spotless-maven-plugin:2.29.0:check (spotless-check) @ seatunnel-config-shade ---
   [INFO] Up-to-date checking enabled
   [INFO] Fingerprint mismatch in the index file. Fallback to an empty index
   [INFO] Sorting file /var/folders/kp/rjp70w_s7qz1lwr6044cq8qh0000gn/T/pom1536588592126316592.xml
   [INFO] Pom file is already sorted, exiting
   [INFO] 
   [INFO] --- maven-remote-resources-plugin:3.1.0:process (process-resource-bundles) @ seatunnel-config-shade ---
   [INFO] Preparing remote bundle org.apache.apache.resources:apache-jar-resource-bundle:1.5
   [INFO] Copying 3 resources from 1 bundle.
   [INFO] 
   [INFO] --- maven-resources-plugin:3.3.1:resources (default-resources) @ seatunnel-config-shade ---
   [INFO] skip non existing resourceDirectory /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-shade/src/main/resources
   [INFO] Copying 3 resources from target/maven-shared-archive-resources to target/classes
   [INFO] 
   [INFO] --- flatten-maven-plugin:1.3.0:flatten (flatten) @ seatunnel-config-shade ---
   [INFO] Generating flattened POM of project org.apache.seatunnel:seatunnel-config-shade:jar:3.0.0-SNAPSHOT...
   [INFO] 
   [INFO] --- maven-compiler-plugin:3.10.1:compile (default-compile) @ seatunnel-config-shade ---
   [INFO] Changes detected - recompiling the module!
   [INFO] Compiling 10 source files to /private/tmp/seatunnel-11271-fix/seatunnel-config/seatunnel-config-shade/target/classes
   [INFO] -------------------------------------------------------------
   [ERROR] COMPILATION ERROR : 
   [INFO] -------------------------------------------------------------
   [ERROR] java.lang.ExceptionInInitializerError
   	at lombok.javac.apt.LombokProcessor.placePostCompileAndDontMakeForceRoundDummiesHook(LombokProcessor.java:174)
   	at lombok.javac.apt.LombokProcessor.init(LombokProcessor.java:96)
   	at lombok.core.AnnotationProcessor$JavacDescriptor.want(AnnotationProcessor.java:160)
   	at lombok.core.AnnotationProcessor.init(AnnotationProcessor.java:213)
   	at lombok.launch.AnnotationProcessorHider$AnnotationProcessor.init(AnnotationProcessor.java:64)
   	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment$ProcessorState.<init>(JavacProcessingEnvironment.java:633)
   	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment$DiscoveredProcessors$ProcessorStateIterator.next(JavacProcessingEnvironment.java:757)
   	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment.discoverAndRunProcs(JavacProcessingEnvironment.java:852)
   	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment$Round.run(JavacProcessingEnvironment.java:1183)
   	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment.doProcessing(JavacProcessingEnvironment.java:1296)
   	at jdk.compiler/com.sun.tools.javac.main.JavaCompiler.processAnnotations(JavaCompiler.java:1276)
   	at jdk.compiler/com.sun.tools.javac.main.JavaCompiler.compile(JavaCompiler.java:952)
   	at jdk.compiler/com.sun.tools.javac.main.Main.compile(Main.java:319)
   	at jdk.compiler/com.sun.tools.javac.main.Main.compile(Main.java:178)
   	at jdk.compiler/com.sun.tools.javac.Main.compile(Main.java:84)
   	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
   	at java.base/java.lang.reflect.Method.invoke(Method.java:565)
   	at org.codehaus.plexus.compiler.javac.JavacCompiler.compileInProcess0(JavacCompiler.java:694)
   	at org.codehaus.plexus.compiler.javac.JavacCompiler.compileInProcessWithProperClassloader(JavacCompiler.java:675)
   	at org.codehaus.plexus.compiler.javac.JavacCompiler.compileInProcess(JavacCompiler.java:664)
   	at org.codehaus.plexus.compiler.javac.JavacCompiler.performCompile(JavacCompiler.java:186)
   	at org.apache.maven.plugin.compiler.AbstractCompilerMojo.execute(AbstractCompilerMojo.java:1209)
   	at org.apache.maven.plugin.compiler.CompilerMojo.execute(CompilerMojo.java:198)
   	at org.apache.maven.plugin.DefaultBuildPluginManager.executeMojo(DefaultBuildPluginManager.java:137)
   	at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:210)
   	at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:156)
   	at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:148)
   	at org.apache.maven.lifecycle.internal.LifecycleModuleBuilder.buildProject(LifecycleModuleBuilder.java:117)
   	at org.apache.maven.lifecycle.internal.LifecycleModuleBuilder.buildProject(LifecycleModuleBuilder.java:81)
   	at org.apache.maven.lifecycle.internal.builder.singlethreaded.SingleThreadedBuilder.build(SingleThreadedBuilder.java:56)
   	at org.apache.maven.lifecycle.internal.LifecycleStarter.execute(LifecycleStarter.java:128)
   	at org.apache.maven.DefaultMaven.doExecute(DefaultMaven.java:305)
   	at org.apache.maven.DefaultMaven.doExecute(DefaultMaven.java:192)
   	at org.apache.maven.DefaultMaven.execute(DefaultMaven.java:105)
   	at org.apache.maven.cli.MavenCli.execute(MavenCli.java:972)
   	at org.apache.maven.cli.MavenCli.doMain(MavenCli.java:293)
   	at org.apache.maven.cli.MavenCli.main(MavenCli.java:196)
   	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
   	at java.base/java.lang.reflect.Method.invoke(Method.java:565)
   	at org.codehaus.plexus.classworlds.launcher.Launcher.launchEnhanced(Launcher.java:282)
   	at org.codehaus.plexus.classworlds.launcher.Launcher.launch(Launcher.java:225)
   	at org.codehaus.plexus.classworlds.launcher.Launcher.mainWithExitCode(Launcher.java:406)
   	at org.codehaus.plexus.classworlds.launcher.Launcher.main(Launcher.java:347)
   	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
   	at java.base/java.lang.reflect.Method.invoke(Method.java:565)
   	at org.apache.maven.wrapper.BootstrapMainStarter.start(BootstrapMainStarter.java:47)
   	at org.apache.maven.wrapper.WrapperExecutor.execute(WrapperExecutor.java:156)
   	at org.apache.maven.wrapper.MavenWrapperMain.main(MavenWrapperMain.java:72)
   Caused by: java.lang.NoSuchFieldException: com.sun.tools.javac.code.TypeTag :: UNKNOWN
   	at lombok.permit.Permit.getField(Permit.java:144)
   	at lombok.javac.JavacTreeMaker$SchroedingerType.getFieldCached(JavacTreeMaker.java:171)
   	at lombok.javac.JavacTreeMaker$TypeTag.typeTag(JavacTreeMaker.java:259)
   	at lombok.javac.Javac.<clinit>(Javac.java:187)
   	... 48 more
   
   [INFO] 1 error
   [INFO] -------------------------------------------------------------
   [INFO] ------------------------------------------------------------------------
   [INFO] Reactor Summary for SeaTunnel : 3.0.0-SNAPSHOT:
   [INFO] 
   [INFO] SeaTunnel : ........................................ SUCCESS [  0.727 s]
   [INFO] SeaTunnel : Config : ............................... SUCCESS [  0.040 s]
   [INFO] SeaTunnel : Config : Base .......................... SUCCESS [  0.449 s]
   [INFO] SeaTunnel : Config : Shade ......................... FAILURE [  0.662 s]
   [INFO] SeaTunnel : Common ................................. SKIPPED
   [INFO] SeaTunnel : Api .................................... SKIPPED
   [INFO] SeaTunnel : Connectors V2 : ........................ SKIPPED
   [INFO] SeaTunnel : Connectors V2 : Common ................. SKIPPED
   [INFO] SeaTunnel : Formats : .............................. SKIPPED
   [INFO] SeaTunnel : Formats : Compatible Debezium Json ..... SKIPPED
   [INFO] SeaTunnel : Connectors V2 : CDC : .................. SKIPPED
   [INFO] SeaTunnel : Connectors V2 : CDC : Base ............. SKIPPED
   [INFO] ------------------------------------------------------------------------
   [INFO] BUILD FAILURE
   [INFO] ------------------------------------------------------------------------
   [INFO] Total time:  2.339 s
   [INFO] Finished at: 2026-08-24T19:54:18+08:00
   [INFO] ------------------------------------------------------------------------
   [ERROR] Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.10.1:compile (default-compile) on project seatunnel-config-shade: Compilation failure
   [ERROR] java.lang.ExceptionInInitializerError
   [ERROR] 	at lombok.javac.apt.LombokProcessor.placePostCompileAndDontMakeForceRoundDummiesHook(LombokProcessor.java:174)
   [ERROR] 	at lombok.javac.apt.LombokProcessor.init(LombokProcessor.java:96)
   [ERROR] 	at lombok.core.AnnotationProcessor$JavacDescriptor.want(AnnotationProcessor.java:160)
   [ERROR] 	at lombok.core.AnnotationProcessor.init(AnnotationProcessor.java:213)
   [ERROR] 	at lombok.launch.AnnotationProcessorHider$AnnotationProcessor.init(AnnotationProcessor.java:64)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment$ProcessorState.<init>(JavacProcessingEnvironment.java:633)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment$DiscoveredProcessors$ProcessorStateIterator.next(JavacProcessingEnvironment.java:757)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment.discoverAndRunProcs(JavacProcessingEnvironment.java:852)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment$Round.run(JavacProcessingEnvironment.java:1183)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.processing.JavacProcessingEnvironment.doProcessing(JavacProcessingEnvironment.java:1296)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.main.JavaCompiler.processAnnotations(JavaCompiler.java:1276)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.main.JavaCompiler.compile(JavaCompiler.java:952)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.main.Main.compile(Main.java:319)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.main.Main.compile(Main.java:178)
   [ERROR] 	at jdk.compiler/com.sun.tools.javac.Main.compile(Main.java:84)
   [ERROR] 	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
   [ERROR] 	at java.base/java.lang.reflect.Method.invoke(Method.java:565)
   [ERROR] 	at org.codehaus.plexus.compiler.javac.JavacCompiler.compileInProcess0(JavacCompiler.java:694)
   [ERROR] 	at org.codehaus.plexus.compiler.javac.JavacCompiler.compileInProcessWithProperClassloader(JavacCompiler.java:675)
   [ERROR] 	at org.codehaus.plexus.compiler.javac.JavacCompiler.compileInProcess(JavacCompiler.java:664)
   [ERROR] 	at org.codehaus.plexus.compiler.javac.JavacCompiler.performCompile(JavacCompiler.java:186)
   [ERROR] 	at org.apache.maven.plugin.compiler.AbstractCompilerMojo.execute(AbstractCompilerMojo.java:1209)
   [ERROR] 	at org.apache.maven.plugin.compiler.CompilerMojo.execute(CompilerMojo.java:198)
   [ERROR] 	at org.apache.maven.plugin.DefaultBuildPluginManager.executeMojo(DefaultBuildPluginManager.java:137)
   [ERROR] 	at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:210)
   [ERROR] 	at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:156)
   [ERROR] 	at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:148)
   [ERROR] 	at org.apache.maven.lifecycle.internal.LifecycleModuleBuilder.buildProject(LifecycleModuleBuilder.java:117)
   [ERROR] 	at org.apache.maven.lifecycle.internal.LifecycleModuleBuilder.buildProject(LifecycleModuleBuilder.java:81)
   [ERROR] 	at org.apache.maven.lifecycle.internal.builder.singlethreaded.SingleThreadedBuilder.build(SingleThreadedBuilder.java:56)
   [ERROR] 	at org.apache.maven.lifecycle.internal.LifecycleStarter.execute(LifecycleStarter.java:128)
   [ERROR] 	at org.apache.maven.DefaultMaven.doExecute(DefaultMaven.java:305)
   [ERROR] 	at org.apache.maven.DefaultMaven.doExecute(DefaultMaven.java:192)
   [ERROR] 	at org.apache.maven.DefaultMaven.execute(DefaultMaven.java:105)
   [ERROR] 	at org.apache.maven.cli.MavenCli.execute(MavenCli.java:972)
   [ERROR] 	at org.apache.maven.cli.MavenCli.doMain(MavenCli.java:293)
   [ERROR] 	at org.apache.maven.cli.MavenCli.main(MavenCli.java:196)
   [ERROR] 	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
   [ERROR] 	at java.base/java.lang.reflect.Method.invoke(Method.java:565)
   [ERROR] 	at org.codehaus.plexus.classworlds.launcher.Launcher.launchEnhanced(Launcher.java:282)
   [ERROR] 	at org.codehaus.plexus.classworlds.launcher.Launcher.launch(Launcher.java:225)
   [ERROR] 	at org.codehaus.plexus.classworlds.launcher.Launcher.mainWithExitCode(Launcher.java:406)
   [ERROR] 	at org.codehaus.plexus.classworlds.launcher.Launcher.main(Launcher.java:347)
   [ERROR] 	at java.base/jdk.internal.reflect.DirectMethodHandleAccessor.invoke(DirectMethodHandleAccessor.java:104)
   [ERROR] 	at java.base/java.lang.reflect.Method.invoke(Method.java:565)
   [ERROR] 	at org.apache.maven.wrapper.BootstrapMainStarter.start(BootstrapMainStarter.java:47)
   [ERROR] 	at org.apache.maven.wrapper.WrapperExecutor.execute(WrapperExecutor.java:156)
   [ERROR] 	at org.apache.maven.wrapper.MavenWrapperMain.main(MavenWrapperMain.java:72)
   [ERROR] Caused by: java.lang.NoSuchFieldException: com.sun.tools.javac.code.TypeTag :: UNKNOWN
   [ERROR] 	at lombok.permit.Permit.getField(Permit.java:144)
   [ERROR] 	at lombok.javac.JavacTreeMaker$SchroedingerType.getFieldCached(JavacTreeMaker.java:171)
   [ERROR] 	at lombok.javac.JavacTreeMaker$TypeTag.typeTag(JavacTreeMaker.java:259)
   [ERROR] 	at lombok.javac.Javac.<clinit>(Javac.java:187)
   [ERROR] 	... 48 more
   [ERROR] 
   [ERROR] 
   [ERROR] -> [Help 1]
   [ERROR] 
   [ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
   [ERROR] Re-run Maven using the -X switch to enable full debug logging.
   [ERROR] 
   [ERROR] For more information about the errors and possible solutions, please read the following articles:
   [ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/MojoFailureException
   [ERROR] 
   [ERROR] After correcting the problems, you can resume the build with the command
   [ERROR]   mvn <args> -rf :seatunnel-config-shade\n- [INFO] Scanning for projects...
   [INFO] 
   [INFO] --------------< org.apache.seatunnel:connector-cdc-base >---------------
   [INFO] Building SeaTunnel : Connectors V2 : CDC : Base 3.0.0-SNAPSHOT
   [INFO] --------------------------------[ jar ]---------------------------------
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/connector-common/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:connector-common:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/connector-common/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [WARNING] org.apache.seatunnel:connector-common:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:connector-common:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/connector-common/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-connectors-v2/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel-connectors-v2:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-connectors-v2/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-format-compatible-debezium-json/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel-format-compatible-debezium-json:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-format-compatible-debezium-json/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [WARNING] org.apache.seatunnel:seatunnel-format-compatible-debezium-json:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:seatunnel-format-compatible-debezium-json:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-format-compatible-debezium-json/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-formats/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel-formats:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-formats/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-api/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel-api:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-api/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [WARNING] org.apache.seatunnel:seatunnel-api:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:seatunnel-api:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-api/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-common/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel-common:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-common/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [WARNING] org.apache.seatunnel:seatunnel-common:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:seatunnel-common:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-common/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-config-shade/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel-config-shade:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-config-shade/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [WARNING] org.apache.seatunnel:seatunnel-config-shade:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:seatunnel-config-shade:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-config-shade/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-config/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel-config:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-config/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   Downloading from liaoliao-nexus-mirror: http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-config-base/3.0.0-SNAPSHOT/maven-metadata.xml
   [WARNING] Could not transfer metadata org.apache.seatunnel:seatunnel-config-base:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-config-base/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [WARNING] org.apache.seatunnel:seatunnel-config-base:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:seatunnel-config-base:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-config-base/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-maven-version) @ connector-cdc-base ---
   [INFO] Rule 0: org.apache.maven.enforcer.rules.version.RequireMavenVersion passed
   [INFO] 
   [INFO] --- maven-enforcer-plugin:3.4.1:enforce (enforce-java-version) @ connector-cdc-base ---
   [INFO] Rule 0: org.apache.maven.enforcer.rules.version.RequireJavaVersion passed
   [INFO] 
   [INFO] --- spotless-maven-plugin:2.29.0:check (spotless-check) @ connector-cdc-base ---
   [INFO] Up-to-date checking enabled
   [INFO] Fingerprint mismatch in the index file. Fallback to an empty index
   [INFO] Sorting file /var/folders/kp/rjp70w_s7qz1lwr6044cq8qh0000gn/T/pom4015450550764392209.xml
   [INFO] Pom file is already sorted, exiting
   [INFO] 
   [INFO] --- maven-remote-resources-plugin:3.1.0:process (process-resource-bundles) @ connector-cdc-base ---
   [INFO] Preparing remote bundle org.apache.apache.resources:apache-jar-resource-bundle:1.5
   [INFO] Copying 3 resources from 1 bundle.
   [WARNING] org.apache.seatunnel:seatunnel-connectors-v2:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:seatunnel-connectors-v2:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-connectors-v2/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [WARNING] org.apache.seatunnel:seatunnel:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:seatunnel:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [WARNING] org.apache.seatunnel:seatunnel-formats:3.0.0-SNAPSHOT/maven-metadata.xmlfailed to transfer from http://124.220.77.162:58081/repository/maven-public/ during a previous attempt. This failure was cached in the local repository and resolution will not be reattempted until the update interval of liaoliao-nexus-mirror has elapsed or updates are forced. Original error: Could not transfer metadata org.apache.seatunnel:seatunnel-formats:3.0.0-SNAPSHOT/maven-metadata.xml from/to liaoliao-nexus-mirror (http://124.220.77.162:58081/repository/maven-public/): authentication failed for http://124.220.77.162:58081/repository/maven-public/org/apache/seatunnel/seatunnel-formats/3.0.0-SNAPSHOT/maven-metadata.xml, status: 401 Unauthorized
   [INFO] 
   [INFO] --- maven-resources-plugin:3.3.1:resources (default-resources) @ connector-cdc-base ---
   [INFO] skip non existing resourceDirectory /private/tmp/seatunnel-11271-fix/seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/resources
   [INFO] Copying 3 resources from target/maven-shared-archive-resources to target/classes
   [INFO] 
   [INFO] --- flatten-maven-plugin:1.3.0:flatten (flatten) @ connector-cdc-base ---
   [INFO] Generating flattened POM of project org.apache.seatunnel:connector-cdc-base:jar:3.0.0-SNAPSHOT...
   [INFO] 
   [INFO] --- maven-compiler-plugin:3.10.1:compile (default-compile) @ connector-cdc-base ---
   [INFO] Nothing to compile - all classes are up to date
   [INFO] 
   [INFO] --- maven-resources-plugin:3.3.1:testResources (default-testResources) @ connector-cdc-base ---
   [INFO] Copying 1 resource from src/test/resources to target/test-classes
   [INFO] Copying 3 resources from target/maven-shared-archive-resources to target/test-classes
   [INFO] 
   [INFO] --- maven-compiler-plugin:3.10.1:testCompile (default-testCompile) @ connector-cdc-base ---
   [INFO] Nothing to compile - all classes are up to date
   [INFO] 
   [INFO] --- maven-surefire-plugin:2.22.2:test (default-test) @ connector-cdc-base ---
   [INFO] 
   [INFO] -------------------------------------------------------
   [INFO]  T E S T S
   [INFO] -------------------------------------------------------
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest
   [ERROR] Tests run: 4, Failures: 0, Errors: 4, Skipped: 0, Time elapsed: 0.358 s <<< FAILURE! - in org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest
   [ERROR] testFetchFinishedSnapshotSplitSupportsNextSplitAfterIdChanges  Time elapsed: 0.338 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitSupportsNextSplitAfterIdChanges(IncrementalSourceSplitReaderTest.java:91)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitSupportsNextSplitAfterIdChanges(IncrementalSourceSplitReaderTest.java:91)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitSupportsNextSplitAfterIdChanges(IncrementalSourceSplitReaderTest.java:91)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitSupportsNextSplitAfterIdChanges(IncrementalSourceSplitReaderTest.java:91)
   
   [ERROR] testFetchFinishedSnapshotSplitFailFastWhenCurrentSplitIdIsNull  Time elapsed: 0.007 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitFailFastWhenCurrentSplitIdIsNull(IncrementalSourceSplitReaderTest.java:69)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitFailFastWhenCurrentSplitIdIsNull(IncrementalSourceSplitReaderTest.java:69)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitFailFastWhenCurrentSplitIdIsNull(IncrementalSourceSplitReaderTest.java:69)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitFailFastWhenCurrentSplitIdIsNull(IncrementalSourceSplitReaderTest.java:69)
   
   [ERROR] testCloseClearsState  Time elapsed: 0.005 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testCloseClearsState(IncrementalSourceSplitReaderTest.java:122)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testCloseClearsState(IncrementalSourceSplitReaderTest.java:122)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testCloseClearsState(IncrementalSourceSplitReaderTest.java:122)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testCloseClearsState(IncrementalSourceSplitReaderTest.java:122)
   
   [ERROR] testFetchFinishedSnapshotSplitEmitsFinishedOnlyOnce  Time elapsed: 0.005 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitEmitsFinishedOnlyOnce(IncrementalSourceSplitReaderTest.java:40)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.dialect.DataSourceDialect, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitEmitsFinishedOnlyOnce(IncrementalSourceSplitReaderTest.java:40)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitEmitsFinishedOnlyOnce(IncrementalSourceSplitReaderTest.java:40)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitEmitsFinishedOnlyOnce(IncrementalSourceSplitReaderTest.java:40)
   
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest
   [ERROR] Tests run: 4, Failures: 0, Errors: 4, Skipped: 0, Time elapsed: 0.01 s <<< FAILURE! - in org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest
   [ERROR] testAddSplitsKeepsRestoredSplitWhenDiscoveryReturnsEmpty  Time elapsed: 0 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryReturnsEmpty(IncrementalSourceReaderTest.java:84)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryReturnsEmpty(IncrementalSourceReaderTest.java:84)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryReturnsEmpty(IncrementalSourceReaderTest.java:84)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryReturnsEmpty(IncrementalSourceReaderTest.java:84)
   
   [ERROR] testAddSplitsEnqueuesPrunedRestoredIncrementalSplit  Time elapsed: 0.002 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsEnqueuesPrunedRestoredIncrementalSplit(IncrementalSourceReaderTest.java:60)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsEnqueuesPrunedRestoredIncrementalSplit(IncrementalSourceReaderTest.java:60)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsEnqueuesPrunedRestoredIncrementalSplit(IncrementalSourceReaderTest.java:60)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsEnqueuesPrunedRestoredIncrementalSplit(IncrementalSourceReaderTest.java:60)
   
   [ERROR] testAddSplitsKeepsRestoredSplitWhenDiscoveryFails  Time elapsed: 0.004 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryFails(IncrementalSourceReaderTest.java:107)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryFails(IncrementalSourceReaderTest.java:107)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryFails(IncrementalSourceReaderTest.java:107)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryFails(IncrementalSourceReaderTest.java:107)
   
   [ERROR] testAddSplitsDiscoversCapturedTablesOnlyOncePerBatch  Time elapsed: 0.004 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsDiscoversCapturedTablesOnlyOncePerBatch(IncrementalSourceReaderTest.java:130)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsDiscoversCapturedTablesOnlyOncePerBatch(IncrementalSourceReaderTest.java:130)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsDiscoversCapturedTablesOnlyOncePerBatch(IncrementalSourceReaderTest.java:130)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.IncrementalSourceReaderTest.testAddSplitsDiscoversCapturedTablesOnlyOncePerBatch(IncrementalSourceReaderTest.java:130)
   
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest
   [ERROR] Tests run: 3, Failures: 0, Errors: 3, Skipped: 0, Time elapsed: 0.151 s <<< FAILURE! - in org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest
   [ERROR] testPollSplitRecordsDrainsQueueBeforeBoundedCompletion  Time elapsed: 0.006 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testPollSplitRecordsDrainsQueueBeforeBoundedCompletion(IncrementalSourceStreamFetcherTest.java:414)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testPollSplitRecordsDrainsQueueBeforeBoundedCompletion(IncrementalSourceStreamFetcherTest.java:414)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testPollSplitRecordsDrainsQueueBeforeBoundedCompletion(IncrementalSourceStreamFetcherTest.java:414)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testPollSplitRecordsDrainsQueueBeforeBoundedCompletion(IncrementalSourceStreamFetcherTest.java:414)
   
   [ERROR] testSplitSchemaChangeStream  Time elapsed: 0.004 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testSplitSchemaChangeStream(IncrementalSourceStreamFetcherTest.java:76)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testSplitSchemaChangeStream(IncrementalSourceStreamFetcherTest.java:76)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testSplitSchemaChangeStream(IncrementalSourceStreamFetcherTest.java:76)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testSplitSchemaChangeStream(IncrementalSourceStreamFetcherTest.java:76)
   
   [ERROR] testNeverStopSentinelIsNotSignaledAsBoundedCompletion  Time elapsed: 0.005 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testNeverStopSentinelIsNotSignaledAsBoundedCompletion(IncrementalSourceStreamFetcherTest.java:462)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeResolver, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testNeverStopSentinelIsNotSignaledAsBoundedCompletion(IncrementalSourceStreamFetcherTest.java:462)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testNeverStopSentinelIsNotSignaledAsBoundedCompletion(IncrementalSourceStreamFetcherTest.java:462)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.createFetcher(IncrementalSourceStreamFetcherTest.java:505)
   	at org.apache.seatunnel.connectors.cdc.base.source.reader.external.IncrementalSourceStreamFetcherTest.testNeverStopSentinelIsNotSignaledAsBoundedCompletion(IncrementalSourceStreamFetcherTest.java:462)
   
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.split.IncrementalSplitTest
   [INFO] Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.004 s - in org.apache.seatunnel.connectors.cdc.base.source.split.IncrementalSplitTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.split.state.IncrementalSplitStateTest
   [INFO] Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0 s - in org.apache.seatunnel.connectors.cdc.base.source.split.state.IncrementalSplitStateTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.enumerator.splitter.AbstractJdbcSourceChunkSplitterTest
   [INFO] Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.003 s - in org.apache.seatunnel.connectors.cdc.base.source.enumerator.splitter.AbstractJdbcSourceChunkSplitterTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest
   [ERROR] Tests run: 4, Failures: 0, Errors: 4, Skipped: 0, Time elapsed: 0.019 s <<< FAILURE! - in org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest
   [ERROR] shouldReturnEmptyWhenNoCapturedTablesRemain  Time elapsed: 0.004 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReturnEmptyWhenNoCapturedTablesRemain(IncrementalSplitAssignerTest.java:109)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReturnEmptyWhenNoCapturedTablesRemain(IncrementalSplitAssignerTest.java:109)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReturnEmptyWhenNoCapturedTablesRemain(IncrementalSplitAssignerTest.java:109)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReturnEmptyWhenNoCapturedTablesRemain(IncrementalSplitAssignerTest.java:109)
   
   [ERROR] shouldReassignTablesRestoredFromCheckpointWithTheirWatermark  Time elapsed: 0.004 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReassignTablesRestoredFromCheckpointWithTheirWatermark(IncrementalSplitAssignerTest.java:126)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReassignTablesRestoredFromCheckpointWithTheirWatermark(IncrementalSplitAssignerTest.java:126)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReassignTablesRestoredFromCheckpointWithTheirWatermark(IncrementalSplitAssignerTest.java:126)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReassignTablesRestoredFromCheckpointWithTheirWatermark(IncrementalSplitAssignerTest.java:126)
   
   [ERROR] shouldSkipEmptyBucketsWhenCapturedTablesAreLessThanParallelism  Time elapsed: 0.005 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldSkipEmptyBucketsWhenCapturedTablesAreLessThanParallelism(IncrementalSplitAssignerTest.java:82)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldSkipEmptyBucketsWhenCapturedTablesAreLessThanParallelism(IncrementalSplitAssignerTest.java:82)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldSkipEmptyBucketsWhenCapturedTablesAreLessThanParallelism(IncrementalSplitAssignerTest.java:82)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldSkipEmptyBucketsWhenCapturedTablesAreLessThanParallelism(IncrementalSplitAssignerTest.java:82)
   
   [ERROR] shouldReuseCheckpointedStartupOffsetAfterRestore  Time elapsed: 0.005 s  <<< ERROR!
   org.mockito.exceptions.base.MockitoException: 
   
   Mockito cannot mock this class: interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig.
   
   If you're not sure why you're getting this error, please open an issue on GitHub.
   
   
   Java               : 26
   JVM vendor name    : Oracle Corporation
   JVM vendor version : 26.0.1+8-34
   JVM name           : OpenJDK 64-Bit Server VM
   JVM version        : 26.0.1+8-34
   JVM info           : mixed mode, sharing
   OS name            : Mac OS X
   OS version         : 26.5.2
   
   
   You are seeing this disclaimer because Mockito is configured to create inlined mocks.
   You can learn about inline mocks and their limitations under item #39 of the Mockito class javadoc.
   
   Underlying exception : org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReuseCheckpointedStartupOffsetAfterRestore(IncrementalSplitAssignerTest.java:52)
   Caused by: org.mockito.exceptions.base.MockitoException: Could not modify all classes [interface org.apache.seatunnel.connectors.cdc.base.config.SourceConfig, interface java.io.Serializable]
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReuseCheckpointedStartupOffsetAfterRestore(IncrementalSplitAssignerTest.java:52)
   Caused by: java.lang.IllegalStateException: 
   
   Byte Buddy could not instrument all classes within the mock's type hierarchy
   
   This problem should never occur for javac-compiled classes. This problem has been observed for classes that are:
    - Compiled by older versions of scalac
    - Classes that are part of the Android distribution
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReuseCheckpointedStartupOffsetAfterRestore(IncrementalSplitAssignerTest.java:52)
   Caused by: java.lang.IllegalArgumentException: Java 26 (70) is not supported by the current version of Byte Buddy which officially supports Java 20 (64) - update Byte Buddy or set net.bytebuddy.experimental as a VM property
   	at org.apache.seatunnel.connectors.cdc.base.source.enumerator.IncrementalSplitAssignerTest.shouldReuseCheckpointedStartupOffsetAfterRestore(IncrementalSplitAssignerTest.java:52)
   
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.enumerator.HybridSplitAssignerTest
   [INFO] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.002 s - in org.apache.seatunnel.connectors.cdc.base.source.enumerator.HybridSplitAssignerTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.source.enumerator.SnapshotSplitAssignerTest
   [INFO] Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0 s - in org.apache.seatunnel.connectors.cdc.base.source.enumerator.SnapshotSplitAssignerTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.utils.SourceRecordUtilsTest
   [INFO] Tests run: 12, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.009 s - in org.apache.seatunnel.connectors.cdc.base.utils.SourceRecordUtilsTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.utils.MessageDelayedEventLimiterTest
   [INFO] Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 20.002 s - in org.apache.seatunnel.connectors.cdc.base.utils.MessageDelayedEventLimiterTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeEventFilterTest
   [INFO] Tests run: 18, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.073 s - in org.apache.seatunnel.connectors.cdc.base.schema.SchemaChangeEventFilterTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.schema.AbstractSchemaChangeResolverTest
   [INFO] Tests run: 7, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.003 s - in org.apache.seatunnel.connectors.cdc.base.schema.AbstractSchemaChangeResolverTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.base.debezium.DebeziumAdapterFactoryTest
   [INFO] Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.001 s - in org.apache.seatunnel.connectors.cdc.base.debezium.DebeziumAdapterFactoryTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.debezium.row.SeaTunnelRowDebeziumDeserializationConvertersTest
   [INFO] Tests run: 6, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.007 s - in org.apache.seatunnel.connectors.cdc.debezium.row.SeaTunnelRowDebeziumDeserializationConvertersTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.debezium.row.DebeziumJsonDeserializeSchemaTest
   [INFO] Tests run: 2, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.081 s - in org.apache.seatunnel.connectors.cdc.debezium.row.DebeziumJsonDeserializeSchemaTest
   [INFO] Running org.apache.seatunnel.connectors.cdc.debezium.format.DebeziumJsonFormatTest
   [INFO] Tests run: 1, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.002 s - in org.apache.seatunnel.connectors.cdc.debezium.format.DebeziumJsonFormatTest
   [INFO] Running jdbc.source.JdbcSourceChunkSplitterTest
   19:54:42.292 [main] ERROR org.apache.seatunnel.connectors.cdc.base.source.enumerator.splitter.AbstractJdbcSourceChunkSplitter - Config snapshotSplitColumn get exception in :java.lang.NullPointerException: Cannot invoke "org.apache.seatunnel.connectors.cdc.base.config.JdbcSourceConfig.getSplitColumn()" because "this.sourceConfig" is null
   19:54:42.295 [main] ERROR org.apache.seatunnel.connectors.cdc.base.source.enumerator.splitter.AbstractJdbcSourceChunkSplitter - Config snapshotSplitColumn get exception in :java.lang.NullPointerException: Cannot invoke "org.apache.seatunnel.connectors.cdc.base.config.JdbcSourceConfig.getSplitColumn()" because "this.sourceConfig" is null
   19:54:42.295 [main] ERROR org.apache.seatunnel.connectors.cdc.base.source.enumerator.splitter.AbstractJdbcSourceChunkSplitter - Config snapshotSplitColumn get exception in :java.lang.NullPointerException: Cannot invoke "org.apache.seatunnel.connectors.cdc.base.config.JdbcSourceConfig.getSplitColumn()" because "this.sourceConfig" is null
   [INFO] Tests run: 3, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 0.004 s - in jdbc.source.JdbcSourceChunkSplitterTest
   [INFO] 
   [INFO] Results:
   [INFO] 
   [ERROR] Errors: 
   [ERROR]   IncrementalSplitAssignerTest.shouldReassignTablesRestoredFromCheckpointWithTheirWatermark:126 Mockito
   [ERROR]   IncrementalSplitAssignerTest.shouldReturnEmptyWhenNoCapturedTablesRemain:109 Mockito
   [ERROR]   IncrementalSplitAssignerTest.shouldReuseCheckpointedStartupOffsetAfterRestore:52 Mockito
   [ERROR]   IncrementalSplitAssignerTest.shouldSkipEmptyBucketsWhenCapturedTablesAreLessThanParallelism:82 Mockito
   [ERROR]   IncrementalSourceReaderTest.testAddSplitsDiscoversCapturedTablesOnlyOncePerBatch:130 Mockito
   [ERROR]   IncrementalSourceReaderTest.testAddSplitsEnqueuesPrunedRestoredIncrementalSplit:60 Mockito
   [ERROR]   IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryFails:107 Mockito
   [ERROR]   IncrementalSourceReaderTest.testAddSplitsKeepsRestoredSplitWhenDiscoveryReturnsEmpty:84 Mockito
   [ERROR]   IncrementalSourceSplitReaderTest.testCloseClearsState:122 Mockito 
   Mockito can...
   [ERROR]   IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitEmitsFinishedOnlyOnce:40 Mockito
   [ERROR]   IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitFailFastWhenCurrentSplitIdIsNull:69 Mockito
   [ERROR]   IncrementalSourceSplitReaderTest.testFetchFinishedSnapshotSplitSupportsNextSplitAfterIdChanges:91 Mockito
   [ERROR]   IncrementalSourceStreamFetcherTest.testNeverStopSentinelIsNotSignaledAsBoundedCompletion:462->createFetcher:505 Mockito
   [ERROR]   IncrementalSourceStreamFetcherTest.testPollSplitRecordsDrainsQueueBeforeBoundedCompletion:414->createFetcher:505 Mockito
   [ERROR]   IncrementalSourceStreamFetcherTest.testSplitSchemaChangeStream:76->createFetcher:505 Mockito
   [INFO] 
   [ERROR] Tests run: 81, Failures: 0, Errors: 15, Skipped: 0
   [INFO] 
   [INFO] ------------------------------------------------------------------------
   [INFO] BUILD FAILURE
   [INFO] ------------------------------------------------------------------------
   [INFO] Total time:  23.603 s
   [INFO] Finished at: 2026-08-24T19:54:42+08:00
   [INFO] ------------------------------------------------------------------------
   [ERROR] Failed to execute goal org.apache.maven.plugins:maven-surefire-plugin:2.22.2:test (default-test) on project connector-cdc-base: There are test failures.
   [ERROR] 
   [ERROR] Please refer to /private/tmp/seatunnel-11271-fix/seatunnel-connectors-v2/connector-cdc/connector-cdc-base/target/surefire-reports for the individual test results.
   [ERROR] Please refer to dump files (if any exist) [date].dump, [date]-jvmRun[N].dump and [date].dumpstream.
   [ERROR] -> [Help 1]
   [ERROR] 
   [ERROR] To see the full stack trace of the errors, re-run Maven with the -e switch.
   [ERROR] Re-run Maven using the -X switch to enable full debug logging.
   [ERROR] 
   [ERROR] For more information about the errors and possible solutions, please read the following articles:
   [ERROR] [Help 1] http://cwiki.apache.org/confluence/display/MAVEN/MojoFailureException — 81 tests, 0 failures/errors/skips.\n\nThe new head is ; the project Build has been queued for this head. The old  decision remains visible, but the source-level blocker identified in the latest review is addressed and the requested incompatible-change documentation is now included.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "hutiefang76 (via GitHub)" <gi...@apache.org>.
hutiefang76 commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5394692855

   Added the requested restore-behavior documentation in 7c86f0b47.\n\nThe English and Chinese incompatible-changes entries cover: the affected CDC-base components, filtering restored state for tables removed from the current capture set, the safe fallback when discovery is unavailable or empty, the user-visible restore impact, and the no-config migration guidance.\n\nI kept this documentation-only follow-up separate from the safety fix in e381b8ebe. Please let me know if the wording should be adjusted to match a release-note convention I missed.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "davidzollo (via GitHub)" <gi...@apache.org>.
davidzollo commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3914153153


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/split/IncrementalSplit.java:
##########
@@ -131,4 +135,48 @@ public IncrementalSplit(
         this.checkpointTables = checkpointTables;
         this.historyTableChanges = historyTableChanges;
     }
+
+    /** Returns restored checkpoint state limited to the tables captured by the current job. */
+    public IncrementalSplit pruneTables(Collection<TableId> capturedTables) {
+        Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
+        List<TableId> filteredTableIds =
+                tableIds.stream().filter(capturedTableSet::contains).collect(Collectors.toList());
+        List<CompletedSnapshotSplitInfo> filteredCompletedSnapshotSplitInfos =
+                completedSnapshotSplitInfos.stream()
+                        .filter(info -> capturedTableSet.contains(info.getTableId()))
+                        .collect(Collectors.toList());
+        List<CatalogTable> filteredCheckpointTables =
+                checkpointTables == null
+                        ? null
+                        : checkpointTables.stream()
+                                .filter(
+                                        table ->
+                                                capturedTableSet.contains(

Review Comment:
   Resolved on the current head (merge commit ee2a1424199, and re-verified after my push at 4bd75ee8128).
   
   You're right that the generic `new TableId(database, schema, table)` reconstruction from `TablePath` doesn't work for Db2, since Db2's Debezium identifiers always use an empty catalog (`new TableId("", schema, table)`).
   
   This was already fixed exactly along the lines you suggested, in commit 7df41cf5d9021a9d00ec05e4f7dda8321ebd80ea:
   - `DataSourceDialect.toTableId(TablePath)` — new default method on the dialect interface, generic `(database, schema, table)` mapping.
   - `Db2Dialect.toTableId(TablePath)` — overridden to return `new TableId("", schema, table)`, matching Db2's empty-catalog format.
   - `IncrementalSplit.pruneTables(Collection<TableId>, Function<TablePath, TableId> tableIdConverter)` — now takes the converter as a parameter instead of hardcoding the generic constructor.
   - `IncrementalSourceReader` calls `incrementalSplit.pruneTables(capturedTables, dataSourceDialect::toTableId)`, so the conversion always goes through the dialect.
   - `IncrementalSplitTest#testPruneTablesUsesDialectSpecificTableIdConverter` and `Db2IncrementalSourceFactoryTest` cover the Db2 empty-catalog case explicitly.
   
   Dismissing the review as resolved. Thanks for the detailed repro.



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "davidzollo (via GitHub)" <gi...@apache.org>.
davidzollo merged PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on code in PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#discussion_r3899617577


##########
seatunnel-connectors-v2/connector-cdc/connector-cdc-base/src/main/java/org/apache/seatunnel/connectors/cdc/base/source/split/IncrementalSplit.java:
##########
@@ -131,4 +135,48 @@ public IncrementalSplit(
         this.checkpointTables = checkpointTables;
         this.historyTableChanges = historyTableChanges;
     }
+
+    /** Returns restored checkpoint state limited to the tables captured by the current job. */
+    public IncrementalSplit pruneTables(Collection<TableId> capturedTables) {
+        Set<TableId> capturedTableSet = new HashSet<>(capturedTables);
+        List<TableId> filteredTableIds =
+                tableIds.stream().filter(capturedTableSet::contains).collect(Collectors.toList());
+        List<CompletedSnapshotSplitInfo> filteredCompletedSnapshotSplitInfos =
+                completedSnapshotSplitInfos.stream()
+                        .filter(info -> capturedTableSet.contains(info.getTableId()))
+                        .collect(Collectors.toList());
+        List<CatalogTable> filteredCheckpointTables =
+                checkpointTables == null
+                        ? null
+                        : checkpointTables.stream()
+                                .filter(
+                                        table ->
+                                                capturedTableSet.contains(

Review Comment:
   @davidzollo Sounds good, thanks for looping @hutiefang76 in directly.
   
   For the record so nothing gets lost: I already confirmed @nzw921rx's Db2 concern is real in my reply above and I've raised it as a High-severity blocker in my latest full review (`pruneTables()` rebuilds a generic `TableId` straight from `CatalogTable#getTablePath()` instead of going through Db2's dialect-specific normalization, so `capturedTableSet.contains(...)` never matches for Db2 and every incremental-phase restore with checkpoint tables gets wiped). No fix has landed on the current head yet, so this is still open.
   
   @hutiefang76 the fix direction @nzw921rx sketched (a `JdbcDataSourceDialect#toTableId(TablePath)` default overridden by `Db2Dialect`, threaded into `pruneTables()`/`IncrementalSourceReader`) looks like the right shape to me as well - it reuses the same normalization `Db2Dialect` already applies elsewhere rather than adding a Db2-only special case inside `pruneTables()`. Happy to take another pass on this as soon as it's pushed.
   



-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-4968426701

   Rechecked the same unchanged head `4496c6e48869` after the CI / merge-gate fact changed.
   
   ## Runtime path rechecked
   ```text
   CDC job restores from checkpoint
     -> restored incremental split reaches IncrementalSourceReader.addSplits(...)
         -> split.asIncrementalSplit()
         -> hasRestoredCheckpointMetadata(...)
             -> checkpointDataType / checkpointTables / historyTableChanges present
         -> dataSourceDialect.discoverDataCollections(sourceConfig)
             -> current captured table set from the latest job config
         -> IncrementalSplit.pruneTables(capturedTables)
             -> prune tableIds
             -> prune completedSnapshotSplitInfos
             -> prune checkpointTables
             -> prune historyTableChanges
         -> if no tables remain: skip the restored split
         -> otherwise: add the pruned split to unfinishedSplits
   ```
   
   ## Merge conclusion
   ### Conclusion: source blocker cleared from Daniel's side
   
   1. Blocking items
   - None from Daniel's source review on this unchanged head.
   
   2. Current gate
   - The current CI signal is now green: https://github.com/apache/seatunnel/actions/runs/28751620088/job/85251635434
   - Daniel only has `READ` permission on this repository, so GitHub may still show earlier review state and a write-capable maintainer may still need to handle the final approval / dismissal step.
   
   I do not see a reopened code-side blocker from Daniel on the current head.
   


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "DanielLeens (via GitHub)" <gi...@apache.org>.
DanielLeens commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5141521937

   Thanks @nielifeng for adding the runtime verification details. That evidence is helpful and matches the source-level path Daniel rechecked earlier.
   
   From Daniel's side, I still do not see a reopened blocker on this unchanged head: the restored incremental split is pruned against the current captured table set before it is added back to the reader queue, and your restore/run verification covers the key user scenario. The remaining step is the normal maintainer review / merge gate handling if GitHub still requires it.
   


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


Re: [PR] [Bug][Connector-V2][CDC] Prune removed tables from restored incremental splits [seatunnel]

Posted by "hutiefang76 (via GitHub)" <gi...@apache.org>.
hutiefang76 commented on PR #11271:
URL: https://github.com/apache/seatunnel/pull/11271#issuecomment-5268695639

   Preserved the deprecated `checkpointDataType` across `pruneTables()` and added regression coverage for both legacy-checkpoint paths: one table removed and no tables removed.
   
   I first ran the new tests against the previous implementation; both failed with `expected STRING but was null`. After the fix:
   
   - `IncrementalSplitTest`: 3 tests, 0 failures/errors
   - `IncrementalSplitTest,IncrementalSourceReaderTest`: 5 tests, 0 failures/errors
   - full `connector-cdc-base` on current `dev`: 68 tests, 0 failures/errors
   - Spotless passed
   
   I also rebased the PR onto the latest `dev` so the next CI run includes the upstream E2E stabilization changes.


-- 
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: commits-unsubscribe@seatunnel.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org