NIFI-554: Catch FileNotFoundException and handle it when creating IndexSearcher

This commit is contained in:
Mark Payne 2015-04-29 08:31:13 -04:00
parent 57e78bf723
commit 20831c87fc
2 changed files with 85 additions and 78 deletions

View File

@ -699,7 +699,7 @@ public class PersistentProvenanceRepository implements ProvenanceEventRepository
if (bytesWrittenSinceRollover.get() >= configuration.getMaxEventFileCapacity()) {
try {
rollover(false);
} catch (IOException e) {
} catch (final IOException e) {
logger.error("Failed to Rollover Provenance Event Repository file due to {}", e.toString());
logger.error("", e);
eventReporter.reportEvent(Severity.ERROR, EVENT_CATEGORY, "Failed to Rollover Provenance Event Repository file due to " + e.toString());
@ -1001,7 +1001,7 @@ public class PersistentProvenanceRepository implements ProvenanceEventRepository
if (fileRolledOver == null) {
return;
}
File file = fileRolledOver;
final File file = fileRolledOver;
// update our map of id to Path
// need lock to update the map, even though it's an AtomicReference, AtomicReference allows those doing a
@ -1010,7 +1010,7 @@ public class PersistentProvenanceRepository implements ProvenanceEventRepository
writeLock.lock();
try {
final Long fileFirstEventId = Long.valueOf(LuceneUtil.substringBefore(fileRolledOver.getName(), "."));
SortedMap<Long, Path> newIdToPathMap = new TreeMap<>(new PathMapComparator());
final SortedMap<Long, Path> newIdToPathMap = new TreeMap<>(new PathMapComparator());
newIdToPathMap.putAll(idToPathMap.get());
newIdToPathMap.put(fileFirstEventId, file.toPath());
idToPathMap.set(newIdToPathMap);
@ -1452,11 +1452,11 @@ public class PersistentProvenanceRepository implements ProvenanceEventRepository
try (final DirectoryReader directoryReader = DirectoryReader.open(FSDirectory.open(indexDirectory))) {
final IndexSearcher searcher = new IndexSearcher(directoryReader);
TopDocs topDocs = searcher.search(luceneQuery, 10000000);
final TopDocs topDocs = searcher.search(luceneQuery, 10000000);
logger.info("For {}, Top Docs has {} hits; reading Lucene results", indexDirectory, topDocs.scoreDocs.length);
if (topDocs.totalHits > 0) {
for (ScoreDoc scoreDoc : topDocs.scoreDocs) {
for (final ScoreDoc scoreDoc : topDocs.scoreDocs) {
final int docId = scoreDoc.doc;
final Document d = directoryReader.document(docId);
localScoreDocs.add(d);
@ -1880,7 +1880,7 @@ public class PersistentProvenanceRepository implements ProvenanceEventRepository
}
try {
final Set<ProvenanceEventRecord> matchingRecords = LineageQuery.computeLineageForFlowFiles(PersistentProvenanceRepository.this, indexDir, null, flowFileUuids);
final Set<ProvenanceEventRecord> matchingRecords = LineageQuery.computeLineageForFlowFiles(PersistentProvenanceRepository.this, indexManager, indexDir, null, flowFileUuids);
final StandardLineageResult result = submission.getResult();
result.update(matchingRecords);

View File

@ -19,26 +19,23 @@ package org.apache.nifi.provenance.lucene;
import static java.util.Objects.requireNonNull;
import java.io.File;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.util.Collection;
import java.util.Collections;
import java.util.Set;
import java.util.concurrent.TimeUnit;
import java.util.concurrent.atomic.AtomicInteger;
import org.apache.nifi.provenance.PersistentProvenanceRepository;
import org.apache.nifi.provenance.ProvenanceEventRecord;
import org.apache.nifi.provenance.SearchableFields;
import org.apache.lucene.index.DirectoryReader;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.Term;
import org.apache.lucene.search.BooleanClause.Occur;
import org.apache.lucene.search.BooleanQuery;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.TermQuery;
import org.apache.lucene.search.TopDocs;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.FSDirectory;
import org.apache.nifi.provenance.PersistentProvenanceRepository;
import org.apache.nifi.provenance.ProvenanceEventRecord;
import org.apache.nifi.provenance.SearchableFields;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
@ -48,7 +45,7 @@ public class LineageQuery {
public static final int MAX_LINEAGE_UUIDS = 100;
private static final Logger logger = LoggerFactory.getLogger(LineageQuery.class);
public static Set<ProvenanceEventRecord> computeLineageForFlowFiles(final PersistentProvenanceRepository repo, final File indexDirectory,
public static Set<ProvenanceEventRecord> computeLineageForFlowFiles(final PersistentProvenanceRepository repo, final IndexManager indexManager, final File indexDirectory,
final String lineageIdentifier, final Collection<String> flowFileUuids) throws IOException {
if (requireNonNull(flowFileUuids).size() > MAX_LINEAGE_UUIDS) {
throw new IllegalArgumentException(String.format("Cannot compute lineage for more than %s FlowFiles. This lineage contains %s.", MAX_LINEAGE_UUIDS, flowFileUuids.size()));
@ -58,11 +55,10 @@ public class LineageQuery {
throw new IllegalArgumentException("Must specify either Lineage Identifier or FlowFile UUIDs to compute lineage");
}
try (final Directory fsDir = FSDirectory.open(indexDirectory);
final IndexReader indexReader = DirectoryReader.open(fsDir)) {
final IndexSearcher searcher = new IndexSearcher(indexReader);
final IndexSearcher searcher;
try {
searcher = indexManager.borrowIndexSearcher(indexDirectory);
try {
// Create a query for all Events related to the FlowFiles of interest. We do this by adding all ID's as
// "SHOULD" clauses and then setting the minimum required to 1.
final BooleanQuery flowFileIdQuery;
@ -98,12 +94,23 @@ public class LineageQuery {
final long searchEnd = System.nanoTime();
final DocsReader docsReader = new DocsReader(repo.getConfiguration().getStorageDirectories());
final Set<ProvenanceEventRecord> recs = docsReader.read(uuidQueryTopDocs, indexReader, repo.getAllLogFiles(), new AtomicInteger(0), Integer.MAX_VALUE);
final Set<ProvenanceEventRecord> recs = docsReader.read(uuidQueryTopDocs, searcher.getIndexReader(), repo.getAllLogFiles(), new AtomicInteger(0), Integer.MAX_VALUE);
final long readDocsEnd = System.nanoTime();
logger.debug("Finished Lineage Query; Lucene search took {} millis, reading records took {} millis",
TimeUnit.NANOSECONDS.toMillis(searchEnd - searchStart), TimeUnit.NANOSECONDS.toMillis(readDocsEnd - searchEnd));
return recs;
} finally {
indexManager.returnIndexSearcher(indexDirectory, searcher);
}
} catch (final FileNotFoundException fnfe) {
// nothing has been indexed yet, or the data has already aged off
logger.warn("Attempted to search Provenance Index {} but could not find the file due to {}", indexDirectory, fnfe);
if ( logger.isDebugEnabled() ) {
logger.warn("", fnfe);
}
return Collections.emptySet();
}
}