Skip to content

Commit

Permalink
fix(core): skip existing items with null ids in StorageServiceSupport…
Browse files Browse the repository at this point in the history
….fetchAllItemsOptimized (#1279) (#1281)

Even though it seems like the existing filters on items coming back from the database
would make this unnecessary, given that existingItems starts empty and, from what I can
tell, only contains (filtered) items from the database.

We've seen this happen in production though.  With any luck the logging will help us
figure out why.

(cherry picked from commit 162c8fb)

Co-authored-by: David Byron <[email protected]>
  • Loading branch information
mergify[bot] and dbyron-sf committed Jul 1, 2023
1 parent 765d822 commit 4fb2bdd
Show file tree
Hide file tree
Showing 2 changed files with 118 additions and 11 deletions.
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@

import static net.logstash.logback.argument.StructuredArguments.value;

import com.google.common.annotations.VisibleForTesting;
import com.google.common.collect.Lists;
import com.netflix.spectator.api.Counter;
import com.netflix.spectator.api.Registry;
Expand Down Expand Up @@ -517,7 +518,8 @@ private Set<T> fetchAllItems(Set<T> existingItems) {
return result;
}

private Set<T> fetchAllItemsOptimized(Set<T> existingItems) {
@VisibleForTesting
Set<T> fetchAllItemsOptimized(Set<T> existingItems) {
if (existingItems == null) {
existingItems = new HashSet<>();
}
Expand Down Expand Up @@ -545,16 +547,18 @@ private Set<T> fetchAllItemsOptimized(Set<T> existingItems) {
Map<String, T> resultMap = new HashMap<>();
existingItems.forEach(
existingItem -> {
String existingItemId = buildObjectKey(existingItem.getId());
if (deletedItems.containsKey(existingItemId)) {
// item was deleted, skip it
log.debug(
"Item with id {} deleted from the store. Will not add to the result.",
existingItemId);
numRemoved.getAndIncrement();
} else if (!modifiedItems.containsKey(existingItemId)) {
// item was unchanged, add it back as is
resultMap.put(existingItemId, existingItem);
if (isIdNotNull(existingItem)) {
String existingItemId = buildObjectKey(existingItem.getId());
if (deletedItems.containsKey(existingItemId)) {
// item was deleted, skip it
log.debug(
"Item with id {} deleted from the store. Will not add to the result.",
existingItemId);
numRemoved.getAndIncrement();
} else if (!modifiedItems.containsKey(existingItemId)) {
// item was unchanged, add it back as is
resultMap.put(existingItemId, existingItem);
}
}
});

Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,103 @@
/*
* Copyright 2023 Salesforce, Inc.
*
* Licensed 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 com.netflix.spinnaker.front50.model;

import static org.assertj.core.api.Assertions.assertThat;
import static org.mockito.ArgumentMatchers.anyLong;
import static org.mockito.ArgumentMatchers.eq;
import static org.mockito.Mockito.doReturn;
import static org.mockito.Mockito.mock;
import static org.mockito.Mockito.verify;

import com.netflix.spectator.api.NoopRegistry;
import com.netflix.spectator.api.Registry;
import com.netflix.spinnaker.front50.api.model.pipeline.Pipeline;
import com.netflix.spinnaker.front50.config.StorageServiceConfigurationProperties;
import io.github.resilience4j.circuitbreaker.CircuitBreakerRegistry;
import java.util.Collections;
import java.util.HashSet;
import java.util.List;
import java.util.Map;
import java.util.Set;
import org.junit.jupiter.api.Test;
import rx.Scheduler;

class StorageServiceSupportTest {

/** Pipeline is an arbitrary choice of an object type. */
class TestDAO extends StorageServiceSupport<Pipeline> {
public TestDAO(
StorageService service,
Scheduler scheduler,
ObjectKeyLoader objectKeyLoader,
StorageServiceConfigurationProperties.PerObjectType configurationProperties,
Registry registry,
CircuitBreakerRegistry circuitBreakerRegistry) {
super(
ObjectType.PIPELINE,
service,
scheduler,
objectKeyLoader,
configurationProperties,
registry,
circuitBreakerRegistry);
}
}

private StorageService storageService = mock(StorageService.class);

private Scheduler scheduler = mock(Scheduler.class);

private StorageServiceConfigurationProperties.PerObjectType testDAOConfigProperties =
new StorageServiceConfigurationProperties.PerObjectType();

private TestDAO testDAO =
new TestDAO(
storageService,
scheduler,
new DefaultObjectKeyLoader(storageService),
testDAOConfigProperties,
new NoopRegistry(),
CircuitBreakerRegistry.ofDefaults());

@Test
void fetchAllItemsOptimizedWithNullId() {
// If presented with an existing item with a null id, make sure
// fetchAllItemsOptimized ignores it, as opposed to trying to build an
// object key for a null id, which results in a NullPointerException.

// No need for any items from the data store for this
List<Pipeline> modifiedItems = Collections.emptyList();
List<Pipeline> deletedItems = Collections.emptyList();
Map<String, List<Pipeline>> newerItems =
Map.of("not_deleted", modifiedItems, "deleted", deletedItems);
doReturn(newerItems)
.when(storageService)
.loadObjectsNewerThan(eq(ObjectType.PIPELINE), anyLong());

HashSet<Pipeline> existingItems = new HashSet<>();
Pipeline itemWithNullId = new Pipeline();
itemWithNullId.setName("pipeline1");
itemWithNullId.setId(null);
existingItems.add(itemWithNullId);

Set<Pipeline> resultingItems = testDAO.fetchAllItemsOptimized(existingItems);
assertThat(resultingItems).isNotNull();
assertThat(resultingItems).isEmpty();

verify(storageService).loadObjectsNewerThan(eq(ObjectType.PIPELINE), anyLong());
}
}

0 comments on commit 4fb2bdd

Please sign in to comment.