Skip to content

Commit c835284

Browse files
janhoyjanhoy
committed
Precommit
1 parent 881c1ec commit c835284

3 files changed

Lines changed: 25 additions & 59 deletions

File tree

‎solr/core/src/java/org/apache/solr/core/ZkContainer.java‎

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -85,9 +85,10 @@ public class ZkContainer {
8585
public ZkContainer() {}
8686

8787
public void initZooKeeper(final CoreContainer cc, CloudConfig config) {
88+
// zkServerEnabled is set whenever in solrCloud mode ('-c') but no explicit zkHost/ZK_HOST is
89+
// provided.
8890
final boolean zkServerEnabled =
8991
EnvUtils.getPropertyAsBool("solr.zookeeper.server.enabled", false);
90-
// TODO NOCOMMIT - understand when zkServerEnabled is set
9192
boolean zkQuorumNode = false;
9293
if (NodeRoles.MODE_ON.equals(cc.nodeRoles.getRoleMode(NodeRoles.Role.ZOOKEEPER_QUORUM))) {
9394
zkQuorumNode = true;
@@ -130,7 +131,7 @@ public void initZooKeeper(final CoreContainer cc, CloudConfig config) {
130131
if (zookeeperHost == null) {
131132
zookeeperHost = zkServer.getClientString();
132133
}
133-
// TODO NOCOMMIT - should this code go in SolrZkServer to augment or replace its current
134+
// TODO - should this code go in SolrZkServer to augment or replace its current
134135
// capabilities? Doing so
135136
// would definitely keep ZkContainer cleaner...
136137
} else {

‎solr/core/src/test/org/apache/solr/cloud/TestEmbeddedZkQuorum.java‎

Lines changed: 22 additions & 51 deletions
Original file line numberDiff line numberDiff line change
@@ -67,7 +67,6 @@ public static void setupCluster() throws Exception {
6767
cluster =
6868
configureCluster(NUM_NODES).addConfig("conf1", configPath).withEmbeddedZkQuorum().build();
6969
cluster.waitForAllNodes(60);
70-
log.info("Cluster configured with {} nodes", NUM_NODES);
7170
}
7271

7372
@Test
@@ -77,16 +76,12 @@ public void testBasicQuorumFunctionality()
7776
JettySolrRunner node = cluster.getJettySolrRunner(i);
7877
assertTrue("Node " + i + " should be running", node.isRunning());
7978
assertNotNull("Node " + i + " should have a NodeName", node.getNodeName());
80-
if (log.isInfoEnabled()) {
81-
log.info("Node {} is running: {}", i, node.getNodeName());
82-
}
8379
}
8480
}
8581

8682
@Test
8783
public void testCollectionIndexing() throws Exception {
8884
try (CloudSolrClient client = cluster.getSolrClient(COLLECTION_NAME)) {
89-
log.info("Creating collection: {}", COLLECTION_NAME);
9085
CollectionAdminRequest.Create createCmd =
9186
CollectionAdminRequest.createCollection(COLLECTION_NAME, "conf1", 1, 3);
9287
createCmd.process(client);
@@ -145,9 +140,6 @@ public void testQuorumResilienceWithNodeFailure() throws Exception {
145140

146141
try {
147142
privateCluster.waitForAllNodes(60);
148-
log.info(
149-
"Private cluster configured with {} nodes for testQuorumResilienceWithNodeFailure",
150-
NUM_NODES);
151143

152144
// Create collection with replica on each node
153145
CollectionAdminRequest.createCollection(collectionName, "conf1", 1, 3)
@@ -163,7 +155,9 @@ public void testQuorumResilienceWithNodeFailure() throws Exception {
163155
// Stop one node (quorum maintained with 2/3 nodes)
164156
JettySolrRunner stoppedNode = privateCluster.getJettySolrRunner(2);
165157
String stoppedNodeName = stoppedNode.getNodeName();
166-
log.info("Stopping node to test quorum resilience: {}", stoppedNodeName);
158+
if (log.isInfoEnabled()) {
159+
log.info("Stopping node to test quorum resilience: {}", stoppedNodeName);
160+
}
167161
privateCluster.stopJettySolrRunner(stoppedNode);
168162

169163
// Wait for ZK to detect node loss and verify cluster still operational
@@ -175,10 +169,10 @@ public void testQuorumResilienceWithNodeFailure() throws Exception {
175169
"documents while node down",
176170
120,
177171
TimeUnit.SECONDS);
178-
log.info("Cluster operational with 2/3 nodes (quorum maintained)");
172+
if (log.isInfoEnabled()) {
173+
log.info("Starting node {} again and testing functionality", stoppedNodeName);
174+
}
179175

180-
// Restart node with same ports (critical for ZK quorum rejoining)
181-
log.info("Restarting node: {}", stoppedNodeName);
182176
privateCluster.startJettySolrRunner(stoppedNode, true);
183177
privateCluster.waitForNode(stoppedNode, 120);
184178

@@ -187,7 +181,7 @@ public void testQuorumResilienceWithNodeFailure() throws Exception {
187181

188182
// CRITICAL: Wait for collection to become active (replicas up, leader elected)
189183
// before attempting to index documents
190-
privateCluster.waitForActiveCollection(collectionName, 1, 3);
184+
privateCluster.waitForActiveCollection(collectionName, 120, TimeUnit.SECONDS, 1, 3);
191185

192186
privateCluster.waitForDocCount(
193187
collectionName,
@@ -204,19 +198,10 @@ public void testQuorumResilienceWithNodeFailure() throws Exception {
204198
"all documents",
205199
120,
206200
TimeUnit.SECONDS);
207-
208-
log.info(
209-
"Node {} successfully rejoined quorum and cluster is fully operational",
210-
stoppedNodeName);
211201
}
212202
} finally {
213-
// Clean up collection and cluster
214-
try {
215-
CollectionAdminRequest.deleteCollection(collectionName)
216-
.process(privateCluster.getSolrClient());
217-
} catch (Exception e) {
218-
log.warn("Failed to delete collection {}: {}", collectionName, e.getMessage());
219-
}
203+
CollectionAdminRequest.deleteCollection(collectionName)
204+
.process(privateCluster.getSolrClient());
220205
privateCluster.shutdown();
221206
}
222207
}
@@ -250,7 +235,6 @@ public void testQuorumLossAndRecovery() throws Exception {
250235

251236
try {
252237
privateCluster.waitForAllNodes(60);
253-
log.info("Private cluster configured with {} nodes for testQuorumLossAndRecovery", NUM_NODES);
254238

255239
// Create collection with 3 replicas (one on each node) to ensure at least
256240
// one replica survives when we stop 2 nodes
@@ -269,68 +253,55 @@ public void testQuorumLossAndRecovery() throws Exception {
269253
String node1Name = node1.getNodeName();
270254
String node2Name = node2.getNodeName();
271255

272-
log.info("Stopping 2 nodes to lose quorum: {}, {}", node1Name, node2Name);
256+
if (log.isInfoEnabled()) {
257+
log.info("Stopping 2 nodes to lose quorum: {}, {}", node1Name, node2Name);
258+
}
273259
privateCluster.stopJettySolrRunner(node1);
274260
privateCluster.stopJettySolrRunner(node2);
275261

276262
// Wait for ZK to detect quorum loss
277263
privateCluster.waitForLiveNodes(1, 120);
278-
log.info("Quorum lost - only 1/3 nodes remaining");
279264

280265
// Restart both nodes to restore quorum
281-
log.info("Restarting nodes to restore quorum");
266+
if (log.isInfoEnabled()) {
267+
log.info("Restarting nodes to restore quorum");
268+
}
282269
privateCluster.startJettySolrRunner(node1, true);
283270
privateCluster.startJettySolrRunner(node2, true);
284271

285272
// Wait for both nodes to register with ZK (they should appear in live_nodes)
286273
// but we don't require them to be fully recovered immediately
287-
try {
288-
privateCluster.waitForNode(node1, 120);
289-
privateCluster.waitForNode(node2, 120);
290-
log.info("Both nodes registered with ZooKeeper");
291-
} catch (Exception e) {
292-
log.warn(
293-
"One or more nodes failed to fully register: {}. Continuing test to verify basic cluster operation.",
294-
e.getMessage());
295-
}
274+
privateCluster.waitForNode(node1, 120);
275+
privateCluster.waitForNode(node2, 120);
296276
privateCluster.waitForLiveNodes(3, 120);
297277

298278
// CRITICAL: Wait for collection to become active (replicas up, leader elected)
299279
// After catastrophic failure, we need to ensure at least one replica is active
300280
// before attempting operations
301-
log.info("Waiting for collection to become active...");
302281
privateCluster.waitForActiveCollection(collectionName, 120, TimeUnit.SECONDS, 1, 1);
303282

304283
// After catastrophic failure, the cluster should be operational with quorum restored
305284
// even if not all replicas are immediately active
306-
log.info("Verifying cluster can query existing data...");
307285
try {
308286
privateCluster.waitForDocCount(
309287
collectionName, 1, "document after recovery", 120, TimeUnit.SECONDS);
310288

311289
// Verify cluster accepts writes
312-
log.info("Verifying cluster accepts writes...");
313290
indexDocuments(client, 1, 1, "after_recovery");
314291
privateCluster.waitForDocCount(
315292
collectionName, 2, "all documents after recovery", 120, TimeUnit.SECONDS);
316293

317-
log.info("Quorum restored successfully - cluster is operational");
318294
} catch (Exception e) {
319-
log.error(
320-
"Cluster failed to become operational after quorum restoration: {}", e.getMessage());
295+
if (log.isErrorEnabled()) {
296+
log.error("Cluster failed to become operational after quorum restoration");
297+
}
321298
throw e;
322299
}
323-
324-
log.info("Quorum restored and cluster fully operational");
325300
}
326301
} finally {
327302
// Clean up collection and cluster
328-
try {
329-
CollectionAdminRequest.deleteCollection(collectionName)
330-
.process(privateCluster.getSolrClient());
331-
} catch (Exception e) {
332-
log.warn("Failed to delete collection {}: {}", collectionName, e.getMessage());
333-
}
303+
CollectionAdminRequest.deleteCollection(collectionName)
304+
.process(privateCluster.getSolrClient());
334305
privateCluster.shutdown();
335306
}
336307
}

‎solr/test-framework/src/java/org/apache/solr/cloud/MiniSolrCloudCluster.java‎

Lines changed: 0 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -573,12 +573,6 @@ private int[] reservePortPairs(int numPairs) throws IOException {
573573
throw new IOException(
574574
"Unable to find " + numPairs + " available port pairs after " + attempts + " attempts");
575575
}
576-
577-
log.info(
578-
"Successfully reserved {} port pairs in {} attempts: {}",
579-
numPairs,
580-
attempts,
581-
Arrays.toString(ports));
582576
return ports;
583577

584578
} finally {

0 commit comments

Comments
 (0)