stevenzwu commented on code in PR #10484: URL: https://github.com/apache/iceberg/pull/10484#discussion_r1637109731
########## flink/v1.19/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/TriggerManager.java: ########## @@ -0,0 +1,295 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.iceberg.flink.maintenance.operator; + +import java.io.IOException; +import java.util.List; +import java.util.stream.Collectors; +import org.apache.flink.annotation.Internal; +import org.apache.flink.api.common.state.ListState; +import org.apache.flink.api.common.state.ListStateDescriptor; +import org.apache.flink.api.common.state.ValueState; +import org.apache.flink.api.common.state.ValueStateDescriptor; +import org.apache.flink.api.common.typeinfo.TypeInformation; +import org.apache.flink.api.common.typeinfo.Types; +import org.apache.flink.configuration.Configuration; +import org.apache.flink.metrics.Counter; +import org.apache.flink.runtime.state.FunctionInitializationContext; +import org.apache.flink.runtime.state.FunctionSnapshotContext; +import org.apache.flink.streaming.api.TimerService; +import org.apache.flink.streaming.api.checkpoint.CheckpointedFunction; +import org.apache.flink.streaming.api.functions.KeyedProcessFunction; +import org.apache.flink.util.Collector; +import org.apache.iceberg.SerializableTable; +import org.apache.iceberg.flink.TableLoader; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** . */ +@Internal +class TriggerManager extends KeyedProcessFunction<Boolean, TableChange, Trigger> Review Comment: I saw `ProcessFunction#Context` also has `timerService()` available. `getRuntimeContext()` also seems to work for `ProcessFunction` ########## flink/v1.19/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/TagBasedLockFactory.java: ########## @@ -0,0 +1,139 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.iceberg.flink.maintenance.operator; + +import java.util.Map; +import org.apache.flink.annotation.Internal; +import org.apache.iceberg.ManageSnapshots; +import org.apache.iceberg.SnapshotRef; +import org.apache.iceberg.Table; +import org.apache.iceberg.flink.TableLoader; +import org.apache.iceberg.util.Tasks; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** + * Iceberg table {@link ManageSnapshots#createTag(String, long)}/{@link + * ManageSnapshots#removeTag(String)} based lock implementation for {@link TriggerLockFactory}. + */ +@Internal +public class TagBasedLockFactory implements TriggerLockFactory { Review Comment: also how are we going to deal with orphaned locks, where `unlock` didn't happen? ########## flink/v1.19/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/TagBasedLockFactory.java: ########## @@ -0,0 +1,139 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.iceberg.flink.maintenance.operator; + +import java.util.Map; +import org.apache.flink.annotation.Internal; +import org.apache.iceberg.ManageSnapshots; +import org.apache.iceberg.SnapshotRef; +import org.apache.iceberg.Table; +import org.apache.iceberg.flink.TableLoader; +import org.apache.iceberg.util.Tasks; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** + * Iceberg table {@link ManageSnapshots#createTag(String, long)}/{@link + * ManageSnapshots#removeTag(String)} based lock implementation for {@link TriggerLockFactory}. + */ +@Internal +public class TagBasedLockFactory implements TriggerLockFactory { Review Comment: Looking at the `PropertiesUpdate` code, it only provides `set` (not `putIfAbsent`). so maybe it doesn't provide the atomicity. My early comment regarding using tag for lock still holds true though. it doesn't feel very natural. I understand the benefit of using Iceberg construct for no additional dependency. But tag-based locks also have higher overhead (metadata JSON file rewrite) compared to other standard solution like SQL database. Iceberg commit's atomicity guarantee in the end also comes from SQL database. ########## flink/v1.19/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/TagBasedLockFactory.java: ########## @@ -0,0 +1,139 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.iceberg.flink.maintenance.operator; + +import java.util.Map; +import org.apache.flink.annotation.Internal; +import org.apache.iceberg.ManageSnapshots; +import org.apache.iceberg.SnapshotRef; +import org.apache.iceberg.Table; +import org.apache.iceberg.flink.TableLoader; +import org.apache.iceberg.util.Tasks; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** + * Iceberg table {@link ManageSnapshots#createTag(String, long)}/{@link + * ManageSnapshots#removeTag(String)} based lock implementation for {@link TriggerLockFactory}. + */ +@Internal +public class TagBasedLockFactory implements TriggerLockFactory { + private static final Logger LOG = LoggerFactory.getLogger(TagBasedLockFactory.class); + private static final String RUNNING_TAG = "__flink_maintenance_running"; + private static final String RECOVERING_TAG = "__flink_maintenance_recovering"; + private static final int CHANGE_ATTEMPTS = 3; + + private final TableLoader tableLoader; + + public TagBasedLockFactory(TableLoader tableLoader) { + this.tableLoader = tableLoader; + } + + @Override + public TriggerLockFactory.Lock createLock() { + return new Lock(tableLoader, RUNNING_TAG); + } + + @Override + public TriggerLockFactory.Lock createRecoveryLock() { + return new Lock(tableLoader, RECOVERING_TAG); + } + + public static class Lock implements TriggerLockFactory.Lock { + private final Table table; + private final String lockKey; + + public Lock(TableLoader tableLoader, String lockKey) { + tableLoader.open(); + this.table = tableLoader.loadTable(); + this.lockKey = lockKey; + } + + /** + * The lock will be acquired by jobs with creating a new tag. A new empty commit is added for a + * table without snapshots. + * + * @return <code>true</code> if the lock is acquired by this operator + */ + @Override + public boolean tryLock() { + table.refresh(); + Map<String, SnapshotRef> refs = table.refs(); + if (isHeld()) { + LOG.info("Lock is already held by someone: {}", refs.keySet()); + return false; + } + + if (table.currentSnapshot() == null) { + // Create an empty commit + table.newFastAppend().commit(); + LOG.info("Empty table, new empty commit added for using tags"); + } + + try { + Tasks.foreach(1) + .retry(CHANGE_ATTEMPTS) + .stopOnFailure() + .throwFailureWhenFinished() + .run( + unused -> { + table.refresh(); + ManageSnapshots manage = table.manageSnapshots(); + manage.createTag(lockKey, table.currentSnapshot().snapshotId()); + manage.commit(); + LOG.debug("Lock created"); + }); + } catch (Exception e) { Review Comment: what if it is a error (like OutOfMemoryError)? ########## flink/v1.19/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/TriggerManager.java: ########## @@ -0,0 +1,295 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.iceberg.flink.maintenance.operator; + +import java.io.IOException; +import java.util.List; +import java.util.stream.Collectors; +import org.apache.flink.annotation.Internal; +import org.apache.flink.api.common.state.ListState; +import org.apache.flink.api.common.state.ListStateDescriptor; +import org.apache.flink.api.common.state.ValueState; +import org.apache.flink.api.common.state.ValueStateDescriptor; +import org.apache.flink.api.common.typeinfo.TypeInformation; +import org.apache.flink.api.common.typeinfo.Types; +import org.apache.flink.configuration.Configuration; +import org.apache.flink.metrics.Counter; +import org.apache.flink.runtime.state.FunctionInitializationContext; +import org.apache.flink.runtime.state.FunctionSnapshotContext; +import org.apache.flink.streaming.api.TimerService; +import org.apache.flink.streaming.api.checkpoint.CheckpointedFunction; +import org.apache.flink.streaming.api.functions.KeyedProcessFunction; +import org.apache.flink.util.Collector; +import org.apache.iceberg.SerializableTable; +import org.apache.iceberg.flink.TableLoader; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** . */ +@Internal +class TriggerManager extends KeyedProcessFunction<Boolean, TableChange, Trigger> + implements CheckpointedFunction { + private static final Logger LOG = LoggerFactory.getLogger(TriggerManager.class); + + private final TableLoader tableLoader; + private final TriggerLockFactory lockFactory; + private final List<String> taskNames; + private final List<TriggerEvaluator> evaluators; + private final long minFireDelayMs; + private final long lockCheckDelayMs; + private final boolean clearLocks; + private transient Counter rateLimiterTriggeredCounter; + private transient Counter concurrentRunTriggeredCounter; + private transient Counter nothingToTriggerCounter; + private transient List<Counter> triggerCounters; + private transient ValueState<Long> nextEvaluationTime; + private transient ListState<TableChange> accumulatedChanges; + private transient ListState<Long> lastTriggerTimes; + private transient TriggerLockFactory.Lock lock; + private transient TriggerLockFactory.Lock recoveryLock; + private transient boolean isCleanUp = false; + private transient boolean inited = false; + private transient int startsFrom = 0; + + TriggerManager( + TableLoader tableLoader, + TriggerLockFactory lockFactory, + List<String> taskNames, + List<TriggerEvaluator> evaluators, + long minFireDelayMs, + long lockCheckDelayMs, + boolean clearLocks) { + Preconditions.checkNotNull(tableLoader, "Table loader should no be null"); + Preconditions.checkNotNull(lockFactory, "Lock factory should no be null"); + Preconditions.checkArgument( + evaluators != null && !evaluators.isEmpty(), "Evaluators should not be empty"); + Preconditions.checkArgument( + taskNames.size() == evaluators.size(), "Provide a name and evaluator for all of the tasks"); + Preconditions.checkArgument(minFireDelayMs > 0, "Minimum fire delay should be at least 1."); + Preconditions.checkArgument( + lockCheckDelayMs > 0, "Minimum lock delay rate should be at least 1."); + + this.tableLoader = tableLoader; + this.lockFactory = lockFactory; + this.taskNames = taskNames; + this.evaluators = evaluators; + this.minFireDelayMs = minFireDelayMs; + this.lockCheckDelayMs = lockCheckDelayMs; + this.clearLocks = clearLocks; + } + + @Override + public void open(Configuration parameters) throws Exception { + this.rateLimiterTriggeredCounter = + getRuntimeContext() + .getMetricGroup() + .addGroup(MetricConstants.GROUP_KEY, MetricConstants.GROUP_VALUE_DEFAULT) + .counter(MetricConstants.RATE_LIMITER_TRIGGERED); + this.concurrentRunTriggeredCounter = + getRuntimeContext() + .getMetricGroup() + .addGroup(MetricConstants.GROUP_KEY, MetricConstants.GROUP_VALUE_DEFAULT) + .counter(MetricConstants.CONCURRENT_RUN_TRIGGERED); + this.nothingToTriggerCounter = + getRuntimeContext() + .getMetricGroup() + .addGroup(MetricConstants.GROUP_KEY, MetricConstants.GROUP_VALUE_DEFAULT) + .counter(MetricConstants.NOTHING_TO_TRIGGER); + this.triggerCounters = + taskNames.stream() + .map( + name -> + getRuntimeContext() + .getMetricGroup() + .addGroup(MetricConstants.GROUP_KEY, name) + .counter(MetricConstants.TRIGGERED)) + .collect(Collectors.toList()); + + this.nextEvaluationTime = + getRuntimeContext() + .getState(new ValueStateDescriptor<>("triggerManagerNextTriggerTime", Types.LONG)); + this.accumulatedChanges = + getRuntimeContext() + .getListState( + new ListStateDescriptor<>( + "triggerManagerAccumulatedChange", TypeInformation.of(TableChange.class))); + this.lastTriggerTimes = + getRuntimeContext() + .getListState(new ListStateDescriptor<>("triggerManagerLastTriggerTime", Types.LONG)); + } + + @Override + public void snapshotState(FunctionSnapshotContext context) throws Exception { + // Do nothing + } + + @Override + public void initializeState(FunctionInitializationContext context) { + LOG.info("Initializing state restored: {}, clearLocks: {}", context.isRestored(), clearLocks); + this.lock = lockFactory.createLock(); + this.recoveryLock = lockFactory.createRecoveryLock(); + if (context.isRestored()) { + isCleanUp = true; + } else if (clearLocks) { + // Remove old lock if we are not restoring the job + lock.unlock(); + } + } + + @Override + public void processElement(TableChange change, Context ctx, Collector<Trigger> out) + throws Exception { + long current = ctx.timerService().currentProcessingTime(); + Long nextTime = nextEvaluationTime.value(); + init(out, ctx.timerService(), current); Review Comment: oh. you move the init logic here (instead of `initializeState`) due to the need of `currentProcessingTime()`? nit: maybe a bit cleaned to move the method to the first line in this method, e.g. `current` can be calculated inside the init method. ``` init(out, ctx.timerService()) ``` ########## flink/v1.19/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/Trigger.java: ########## @@ -0,0 +1,91 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.iceberg.flink.maintenance.operator; + +import java.util.Objects; +import org.apache.flink.annotation.Internal; +import org.apache.iceberg.SerializableTable; +import org.apache.iceberg.relocated.com.google.common.base.MoreObjects; + +@Internal +class Trigger { + private final long timestamp; + private final SerializableTable table; + private final Integer taskId; + private final boolean isCleanUp; + + private Trigger(long timestamp, SerializableTable table, Integer taskId, boolean isCleanUp) { + this.timestamp = timestamp; + this.table = table; + this.taskId = taskId; Review Comment: I don't fully understand the 2 options. we can postpone this discussion to later PRs when the taskId is actually used. in the 2nd option, why does watermark generator needs to be n instances? ########## flink/v1.19/flink/src/main/java/org/apache/iceberg/flink/maintenance/operator/TriggerManager.java: ########## @@ -0,0 +1,295 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ +package org.apache.iceberg.flink.maintenance.operator; + +import java.io.IOException; +import java.util.List; +import java.util.stream.Collectors; +import org.apache.flink.annotation.Internal; +import org.apache.flink.api.common.state.ListState; +import org.apache.flink.api.common.state.ListStateDescriptor; +import org.apache.flink.api.common.state.ValueState; +import org.apache.flink.api.common.state.ValueStateDescriptor; +import org.apache.flink.api.common.typeinfo.TypeInformation; +import org.apache.flink.api.common.typeinfo.Types; +import org.apache.flink.configuration.Configuration; +import org.apache.flink.metrics.Counter; +import org.apache.flink.runtime.state.FunctionInitializationContext; +import org.apache.flink.runtime.state.FunctionSnapshotContext; +import org.apache.flink.streaming.api.TimerService; +import org.apache.flink.streaming.api.checkpoint.CheckpointedFunction; +import org.apache.flink.streaming.api.functions.KeyedProcessFunction; +import org.apache.flink.util.Collector; +import org.apache.iceberg.SerializableTable; +import org.apache.iceberg.flink.TableLoader; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.apache.iceberg.relocated.com.google.common.collect.Lists; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +/** . */ +@Internal +class TriggerManager extends KeyedProcessFunction<Boolean, TableChange, Trigger> + implements CheckpointedFunction { + private static final Logger LOG = LoggerFactory.getLogger(TriggerManager.class); + + private final TableLoader tableLoader; + private final TriggerLockFactory lockFactory; + private final List<String> taskNames; + private final List<TriggerEvaluator> evaluators; + private final long minFireDelayMs; + private final long lockCheckDelayMs; + private final boolean clearLocks; + private transient Counter rateLimiterTriggeredCounter; + private transient Counter concurrentRunTriggeredCounter; + private transient Counter nothingToTriggerCounter; + private transient List<Counter> triggerCounters; + private transient ValueState<Long> nextEvaluationTime; + private transient ListState<TableChange> accumulatedChanges; + private transient ListState<Long> lastTriggerTimes; + private transient TriggerLockFactory.Lock lock; + private transient TriggerLockFactory.Lock recoveryLock; + private transient boolean isCleanUp = false; + private transient boolean inited = false; + private transient int startsFrom = 0; + + TriggerManager( + TableLoader tableLoader, + TriggerLockFactory lockFactory, + List<String> taskNames, + List<TriggerEvaluator> evaluators, + long minFireDelayMs, + long lockCheckDelayMs, + boolean clearLocks) { + Preconditions.checkNotNull(tableLoader, "Table loader should no be null"); + Preconditions.checkNotNull(lockFactory, "Lock factory should no be null"); + Preconditions.checkArgument( + evaluators != null && !evaluators.isEmpty(), "Evaluators should not be empty"); + Preconditions.checkArgument( + taskNames.size() == evaluators.size(), "Provide a name and evaluator for all of the tasks"); + Preconditions.checkArgument(minFireDelayMs > 0, "Minimum fire delay should be at least 1."); + Preconditions.checkArgument( + lockCheckDelayMs > 0, "Minimum lock delay rate should be at least 1."); + + this.tableLoader = tableLoader; + this.lockFactory = lockFactory; + this.taskNames = taskNames; + this.evaluators = evaluators; + this.minFireDelayMs = minFireDelayMs; + this.lockCheckDelayMs = lockCheckDelayMs; + this.clearLocks = clearLocks; + } + + @Override + public void open(Configuration parameters) throws Exception { + this.rateLimiterTriggeredCounter = + getRuntimeContext() + .getMetricGroup() + .addGroup(MetricConstants.GROUP_KEY, MetricConstants.GROUP_VALUE_DEFAULT) + .counter(MetricConstants.RATE_LIMITER_TRIGGERED); + this.concurrentRunTriggeredCounter = + getRuntimeContext() + .getMetricGroup() + .addGroup(MetricConstants.GROUP_KEY, MetricConstants.GROUP_VALUE_DEFAULT) + .counter(MetricConstants.CONCURRENT_RUN_TRIGGERED); + this.nothingToTriggerCounter = + getRuntimeContext() + .getMetricGroup() + .addGroup(MetricConstants.GROUP_KEY, MetricConstants.GROUP_VALUE_DEFAULT) + .counter(MetricConstants.NOTHING_TO_TRIGGER); + this.triggerCounters = + taskNames.stream() + .map( + name -> + getRuntimeContext() + .getMetricGroup() + .addGroup(MetricConstants.GROUP_KEY, name) + .counter(MetricConstants.TRIGGERED)) + .collect(Collectors.toList()); + + this.nextEvaluationTime = + getRuntimeContext() + .getState(new ValueStateDescriptor<>("triggerManagerNextTriggerTime", Types.LONG)); + this.accumulatedChanges = + getRuntimeContext() + .getListState( + new ListStateDescriptor<>( + "triggerManagerAccumulatedChange", TypeInformation.of(TableChange.class))); + this.lastTriggerTimes = + getRuntimeContext() + .getListState(new ListStateDescriptor<>("triggerManagerLastTriggerTime", Types.LONG)); + } + + @Override + public void snapshotState(FunctionSnapshotContext context) throws Exception { + // Do nothing + } + + @Override + public void initializeState(FunctionInitializationContext context) { + LOG.info("Initializing state restored: {}, clearLocks: {}", context.isRestored(), clearLocks); + this.lock = lockFactory.createLock(); + this.recoveryLock = lockFactory.createRecoveryLock(); + if (context.isRestored()) { + isCleanUp = true; + } else if (clearLocks) { + // Remove old lock if we are not restoring the job + lock.unlock(); + } + } + + @Override + public void processElement(TableChange change, Context ctx, Collector<Trigger> out) + throws Exception { + long current = ctx.timerService().currentProcessingTime(); + Long nextTime = nextEvaluationTime.value(); + init(out, ctx.timerService(), current); + + // Add the new changes to the already accumulated ones + List<TableChange> accumulated = Lists.newArrayList(accumulatedChanges.get()); Review Comment: still not quite following. I thought TriggerManager accumulates `TableChange` and decide if it is time to actually trigger/schedule the maintenance task. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
