kbuci commented on code in PR #18302:
URL: https://github.com/apache/hudi/pull/18302#discussion_r2944533138
##########
hudi-client/hudi-client-common/src/main/java/org/apache/hudi/config/HoodieWriteConfig.java:
##########
@@ -674,6 +675,35 @@ public class HoodieWriteConfig extends HoodieConfig {
.markAdvanced()
.withDocumentation("Number of heartbeat misses, before a writer is
deemed not alive and all pending writes are aborted.");
+ public static final ConfigProperty<Boolean> ROLLBACK_FAILED_CLUSTERING =
ConfigProperty
+ .key("hoodie.rollback.failed.clustering")
+ .defaultValue(false)
+ .withInferFunction(cfg -> {
+ String strategy =
cfg.getStringOrDefault(HoodieLockConfig.WRITE_CONFLICT_RESOLUTION_STRATEGY_CLASS_NAME,
"");
+ if
(PreferWriterConflictResolutionStrategy.class.getName().equals(strategy)) {
+ return Option.of(true);
+ }
+ return Option.empty();
+ })
+ .markAdvanced()
+ .withDocumentation("When enabled, rollback of failed writes (under LAZY
cleaning policy) will also attempt to rollback "
+ + "clustering replacecommit instants whose heartbeat has expired.
This is automatically enabled when using "
+ + "PreferWriterConflictResolutionStrategy. Clustering jobs will
start a heartbeat before scheduling a plan, "
+ + "so that other writers can detect stale/failed clustering
attempts. Note that the same "
+ + "client must be used to schedule, execute, and commit the
clustering instant.");
+
+ public static final ConfigProperty<Long>
ROLLBACK_FAILED_CLUSTERING_WAIT_MINUTES = ConfigProperty
+ .key("hoodie.rollback.failed.clustering.wait.minutes")
+ .defaultValue(60L)
Review Comment:
We wait x minutes until after the requested instant is created, updated doc
to clarify that
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
