/*
* 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.cassandra.db.compaction;
import java.io.File;
import java.io.FileOutputStream;
import java.io.IOException;
import java.util.*;
import com.google.common.annotations.VisibleForTesting;
import com.google.common.base.Predicate;
import com.google.common.base.Predicates;
import com.google.common.collect.ImmutableSortedSet;
import com.google.common.collect.Iterables;
import com.google.common.collect.Sets;
import com.google.common.primitives.Ints;
import org.codehaus.jackson.JsonEncoding;
import org.codehaus.jackson.JsonFactory;
import org.codehaus.jackson.JsonGenerator;
import org.codehaus.jackson.JsonNode;
import org.codehaus.jackson.map.ObjectMapper;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.apache.cassandra.db.ColumnFamilyStore;
import org.apache.cassandra.db.RowPosition;
import org.apache.cassandra.dht.Bounds;
import org.apache.cassandra.dht.Token;
import org.apache.cassandra.io.FSWriteError;
import org.apache.cassandra.io.sstable.SSTable;
import org.apache.cassandra.io.sstable.SSTableReader;
import org.apache.cassandra.io.util.FileUtils;
public class LeveledManifest
{
private static final Logger logger = LoggerFactory.getLogger(LeveledManifest.class);
public static final String EXTENSION = ".json";
/**
* limit the number of L0 sstables we do at once, because compaction bloom filter creation
* uses a pessimistic estimate of how many keys overlap (none), so we risk wasting memory
* or even OOMing when compacting highly overlapping sstables
*/
private static final int MAX_COMPACTING_L0 = 32;
private final ColumnFamilyStore cfs;
private final List<SSTableReader>[] generations;
private final Map<SSTableReader, Integer> sstableGenerations;
private final RowPosition[] lastCompactedKeys;
private final int maxSSTableSizeInBytes;
private LeveledManifest(ColumnFamilyStore cfs, int maxSSTableSizeInMB)
{
this.cfs = cfs;
this.maxSSTableSizeInBytes = maxSSTableSizeInMB * 1024 * 1024;
// allocate enough generations for a PB of data
int n = (int) Math.log10(1000 * 1000 * 1000 / maxSSTableSizeInMB);
generations = new List[n];
lastCompactedKeys = new RowPosition[n];
for (int i = 0; i < generations.length; i++)
{
generations[i] = new ArrayList<SSTableReader>();
lastCompactedKeys[i] = cfs.partitioner.getMinimumToken().minKeyBound();
}
sstableGenerations = new HashMap<SSTableReader, Integer>();
}
static LeveledManifest create(ColumnFamilyStore cfs, int maxSSTableSize)
{
return create(cfs, maxSSTableSize, cfs.getSSTables());
}
public static LeveledManifest create(ColumnFamilyStore cfs, int maxSSTableSize, Iterable<SSTableReader> sstables)
{
LeveledManifest manifest = new LeveledManifest(cfs, maxSSTableSize);
load(cfs, manifest, sstables);
// ensure all SSTables are in the manifest
for (SSTableReader ssTableReader : sstables)
{
if (manifest.levelOf(ssTableReader) < 0)
manifest.add(ssTableReader);
}
for (int i = 1; i < manifest.getAllLevelSize().length; i++)
manifest.repairOverlappingSSTables(i);
return manifest;
}
private static void load(ColumnFamilyStore cfs, LeveledManifest manifest, Iterable<SSTableReader> sstables)
{
File manifestFile = tryGetManifest(cfs);
if (manifestFile == null)
return;
try
{
parseManifest(manifest, sstables, manifestFile);
}
catch (Exception e)
{
logger.debug("Error parsing manifest", e);
File oldFile = new File(manifestFile.getPath().replace(EXTENSION, "-old.json"));
if (oldFile.exists())
{
try
{
parseManifest(manifest, sstables, oldFile);
return;
}
catch (Exception old)
{
logger.debug("Old manifest present but corrupt", old);
}
}
logger.warn("Manifest present but corrupt. Cassandra will re-level {} from scratch", cfs.columnFamily);
}
}
private static void parseManifest(LeveledManifest manifest, Iterable<SSTableReader> sstables, File manifestFile) throws IOException
{
ObjectMapper m = new ObjectMapper();
JsonNode rootNode = m.readValue(manifestFile, JsonNode.class);
JsonNode generations = rootNode.get("generations");
assert generations.isArray();
for (JsonNode generation : generations)
{
int level = generation.get("generation").getIntValue();
JsonNode generationValues = generation.get("members");
for (JsonNode generationValue : generationValues)
{
for (SSTableReader ssTableReader : sstables)
{
if (ssTableReader.descriptor.generation == generationValue.getIntValue())
{
logger.debug("Loading {} at L{}", ssTableReader, level);
manifest.add(ssTableReader, level);
}
}
}
}
}
public synchronized void add(SSTableReader reader)
{
logDistribution();
logger.debug("Adding {} to L0", reader);
add(reader, 0);
serialize();
}
/**
* if the number of SSTables in the current compacted set *by itself* exceeds the target level's
* (regardless of the level's current contents), find an empty level instead
*/
private int skipLevels(int newLevel, Iterable<SSTableReader> added)
{
while (maxBytesForLevel(newLevel) < SSTableReader.getTotalBytes(added)
&& generations[(newLevel + 1)].isEmpty())
{
newLevel++;
}
return newLevel;
}
public synchronized void promote(Iterable<SSTableReader> removed, Iterable<SSTableReader> added)
{
assert !Iterables.isEmpty(removed); // use add() instead of promote when adding new sstables
logDistribution();
if (logger.isDebugEnabled())
logger.debug("Replacing [" + toString(removed) + "]");
// the level for the added sstables is the max of the removed ones,
// plus one if the removed were all on the same level
int minimumLevel = Integer.MAX_VALUE;
int maximumLevel = 0;
for (SSTableReader sstable : removed)
{
int thisLevel = remove(sstable);
assert thisLevel >= 0;
maximumLevel = Math.max(maximumLevel, thisLevel);
minimumLevel = Math.min(minimumLevel, thisLevel);
}
// it's valid to do a remove w/o an add (e.g. on truncate)
if (!added.iterator().hasNext())
return;
int newLevel;
if (minimumLevel == 0 && maximumLevel == 0 && SSTable.getTotalBytes(removed) <= maxSSTableSizeInBytes)
{
// special case for tiny L0 sstables; see CASSANDRA-4341
newLevel = 0;
}
else
{
newLevel = minimumLevel == maximumLevel ? maximumLevel + 1 : maximumLevel;
newLevel = skipLevels(newLevel, added);
assert newLevel > 0;
}
if (logger.isDebugEnabled())
logger.debug("Adding [{}] at L{}", toString(added), newLevel);
lastCompactedKeys[minimumLevel] = SSTable.sstableOrdering.max(added).last;
for (SSTableReader ssTableReader : added)
add(ssTableReader, newLevel);
// Fix overlapping sstables from CASSANDRA-4321/4411
if (newLevel != 0)
repairOverlappingSSTables(newLevel);
serialize();
}
public synchronized void repairOverlappingSSTables(int level)
{
SSTableReader previous = null;
Collections.sort(generations[level], SSTable.sstableComparator);
List<SSTableReader> outOfOrderSSTables = new ArrayList<SSTableReader>();
for (SSTableReader current : generations[level])
{
if (previous != null && current.first.compareTo(previous.last) <= 0)
{
logger.error(String.format("At level %d, %s [%s, %s] overlaps %s [%s, %s]. This is caused by a bug in Cassandra 1.1.0 .. 1.1.3. Sending back to L0. If you have not yet run scrub, you should do so since you may also have rows out-of-order within an sstable",
level, previous, previous.first, previous.last, current, current.first, current.last));
outOfOrderSSTables.add(current);
}
else
{
previous = current;
}
}
if (!outOfOrderSSTables.isEmpty())
{
for (SSTableReader sstable : outOfOrderSSTables)
sendBackToL0(sstable);
serialize();
}
}
public synchronized void replace(Iterable<SSTableReader> removed, Iterable<SSTableReader> added)
{
// replace is for compaction operation that operate on exactly one sstable, with no merging.
// Thus, removed will be exactly one sstable, and added will be 0 or 1.
assert Iterables.size(removed) == 1 : Iterables.size(removed);
assert Iterables.size(added) <= 1 : Iterables.size(added);
logDistribution();
logger.debug("Replacing {} with {}", removed, added);
int level = remove(removed.iterator().next());
if (!Iterables.isEmpty(added))
add(added.iterator().next(), level);
serialize();
}
private synchronized void sendBackToL0(SSTableReader sstable)
{
remove(sstable);
add(sstable, 0);
}
private String toString(Iterable<SSTableReader> sstables)
{
StringBuilder builder = new StringBuilder();
for (SSTableReader sstable : sstables)
{
builder.append(sstable.descriptor.cfname)
.append('-')
.append(sstable.descriptor.generation)
.append("(L")
.append(levelOf(sstable))
.append("), ");
}
return builder.toString();
}
@VisibleForTesting
long maxBytesForLevel(int level)
{
if (level == 0)
return 4L * maxSSTableSizeInBytes;
double bytes = Math.pow(10, level) * maxSSTableSizeInBytes;
if (bytes > Long.MAX_VALUE)
throw new RuntimeException("At most " + Long.MAX_VALUE + " bytes may be in a compaction level; your maxSSTableSize must be absurdly high to compute " + bytes);
return (long) bytes;
}
/**
* @return highest-priority sstables to compact
* If no compactions are necessary, will return an empty list. Never returns null.
*/
public synchronized Collection<SSTableReader> getCompactionCandidates()
{
// LevelDB gives each level a score of how much data it contains vs its ideal amount, and
// compacts the level with the highest score. But this falls apart spectacularly once you
// get behind. Consider this set of levels:
// L0: 988 [ideal: 4]
// L1: 117 [ideal: 10]
// L2: 12 [ideal: 100]
//
// The problem is that L0 has a much higher score (almost 250) than L1 (11), so what we'll
// do is compact a batch of MAX_COMPACTING_L0 sstables with all 117 L1 sstables, and put the
// result (say, 120 sstables) in L1. Then we'll compact the next batch of MAX_COMPACTING_L0,
// and so forth. So we spend most of our i/o rewriting the L1 data with each batch.
//
// If we could just do *all* L0 a single time with L1, that would be ideal. But we can't
// -- see the javadoc for MAX_COMPACTING_L0.
//
// LevelDB's way around this is to simply block writes if L0 compaction falls behind.
// We don't have that luxury.
//
// So instead, we force compacting higher levels first. This may not minimize the number
// of reads done as quickly in the short term, but it minimizes the i/o needed to compact
// optimially which gives us a long term win.
for (int i = generations.length - 1; i >= 0; i--)
{
List<SSTableReader> sstables = generations[i];
if (sstables.isEmpty())
continue; // mostly this just avoids polluting the debug log with zero scores
// we want to calculate score excluding compacting ones
Set<SSTableReader> sstablesInLevel = Sets.newHashSet(sstables);
Set<SSTableReader> remaining = Sets.difference(sstablesInLevel, cfs.getDataTracker().getCompacting());
double score = (double)SSTableReader.getTotalBytes(remaining) / (double)maxBytesForLevel(i);
logger.debug("Compaction score for level {} is {}", i, score);
// L0 gets a special case that if we don't have anything more important to do,
// we'll go ahead and compact if we have more than one sstable
if (score > 1.001 || (i == 0 && sstables.size() > 1))
{
Collection<SSTableReader> candidates = getCandidatesFor(i);
if (logger.isDebugEnabled())
logger.debug("Compaction candidates for L{} are {}", i, toString(candidates));
if (!candidates.isEmpty())
return candidates;
}
}
return Collections.emptyList();
}
public int getLevelSize(int i)
{
return generations.length > i ? generations[i].size() : 0;
}
public synchronized int[] getAllLevelSize()
{
int[] counts = new int[generations.length];
for (int i = 0; i < counts.length; i++)
counts[i] = generations[i].size();
return counts;
}
private void logDistribution()
{
if (logger.isDebugEnabled())
{
for (int i = 0; i < generations.length; i++)
{
if (!generations[i].isEmpty())
{
logger.debug("L{} contains {} SSTables ({} bytes) in {}",
new Object[] {i, generations[i].size(), SSTableReader.getTotalBytes(generations[i]), this});
}
}
}
}
int levelOf(SSTableReader sstable)
{
Integer level = sstableGenerations.get(sstable);
if (level == null)
return -1;
return level.intValue();
}
private int remove(SSTableReader reader)
{
int level = levelOf(reader);
assert level >= 0 : reader + " not present in manifest";
generations[level].remove(reader);
sstableGenerations.remove(reader);
return level;
}
private void add(SSTableReader sstable, int level)
{
assert level < generations.length : "Invalid level " + level + " out of " + (generations.length - 1);
generations[level].add(sstable);
sstableGenerations.put(sstable, Integer.valueOf(level));
}
private static Set<SSTableReader> overlapping(Collection<SSTableReader> candidates, Iterable<SSTableReader> others)
{
assert !candidates.isEmpty();
/*
* Picking each sstable from others that overlap one of the sstable of candidates is not enough
* because you could have the following situation:
* candidates = [ s1(a, c), s2(m, z) ]
* others = [ s3(e, g) ]
* In that case, s2 overlaps none of s1 or s2, but if we compact s1 with s2, the resulting sstable will
* overlap s3, so we must return s3.
*
* Thus, the correct approach is to pick sstables overlapping anything between the first key in all
* the candidate sstables, and the last.
*/
Iterator<SSTableReader> iter = candidates.iterator();
SSTableReader sstable = iter.next();
Token first = sstable.first.token;
Token last = sstable.last.token;
while (iter.hasNext())
{
sstable = iter.next();
first = first.compareTo(sstable.first.token) <= 0 ? first : sstable.first.token;
last = last.compareTo(sstable.last.token) >= 0 ? last : sstable.last.token;
}
return overlapping(first, last, others);
}
@VisibleForTesting
static Set<SSTableReader> overlapping(SSTableReader sstable, Iterable<SSTableReader> others)
{
return overlapping(sstable.first.token, sstable.last.token, others);
}
/**
* @return sstables from @param sstables that contain keys between @param start and @param end, inclusive.
*/
private static Set<SSTableReader> overlapping(Token start, Token end, Iterable<SSTableReader> sstables)
{
assert start.compareTo(end) <= 0;
Set<SSTableReader> overlapped = new HashSet<SSTableReader>();
Bounds<Token> promotedBounds = new Bounds<Token>(start, end);
for (SSTableReader candidate : sstables)
{
Bounds<Token> candidateBounds = new Bounds<Token>(candidate.first.token, candidate.last.token);
if (candidateBounds.intersects(promotedBounds))
overlapped.add(candidate);
}
return overlapped;
}
private static final Predicate<SSTableReader> suspectP = new Predicate<SSTableReader>()
{
public boolean apply(SSTableReader candidate)
{
return candidate.isMarkedSuspect();
}
};
/**
* @return highest-priority sstables to compact for the given level.
* If no compactions are possible (because of concurrent compactions or because some sstables are blacklisted
* for prior failure), will return an empty list. Never returns null.
*/
private Collection<SSTableReader> getCandidatesFor(int level)
{
assert !generations[level].isEmpty();
logger.debug("Choosing candidates for L{}", level);
final Set<SSTableReader> compacting = cfs.getDataTracker().getCompacting();
if (level == 0)
{
// L0 is the dumping ground for new sstables which thus may overlap each other.
//
// We treat L0 compactions specially:
// 1a. add sstables to the candidate set until we have at least maxSSTableSizeInMB
// 1b. prefer choosing older sstables as candidates, to newer ones
// 1c. any L0 sstables that overlap a candidate, will also become candidates
// 2. At most MAX_COMPACTING_L0 sstables from L0 will be compacted at once
// 3. If total candidate size is less than maxSSTableSizeInMB, we won't bother compacting with L1,
// and the result of the compaction will stay in L0 instead of being promoted (see promote())
//
// Note that we ignore suspect-ness of L1 sstables here, since if an L1 sstable is suspect we're
// basically screwed, since we expect all or most L0 sstables to overlap with each L1 sstable.
// So if an L1 sstable is suspect we can't do much besides try anyway and hope for the best.
Set<SSTableReader> candidates = new HashSet<SSTableReader>();
Set<SSTableReader> remaining = new HashSet<SSTableReader>();
Iterables.addAll(remaining, Iterables.filter(generations[0], Predicates.not(suspectP)));
for (SSTableReader sstable : ageSortedSSTables(remaining))
{
if (candidates.contains(sstable))
continue;
for (SSTableReader newCandidate : Sets.union(Collections.singleton(sstable), overlapping(sstable, remaining)))
{
if (!compacting.contains(newCandidate))
{
candidates.add(newCandidate);
remaining.remove(newCandidate);
}
}
if (candidates.size() > MAX_COMPACTING_L0)
{
// limit to only the MAX_COMPACTING_L0 oldest candidates
candidates = new HashSet<SSTableReader>(ageSortedSSTables(candidates).subList(0, MAX_COMPACTING_L0));
break;
}
}
// leave everything in L0 if we didn't end up with a full sstable's worth of data
if (SSTable.getTotalBytes(candidates) > maxSSTableSizeInBytes)
{
// add sstables from L1 that overlap candidates
// if the overlapping ones are already busy in a compaction, leave it out.
// TODO try to find a set of L0 sstables that only overlaps with non-busy L1 sstables
candidates = Sets.union(candidates, overlapping(candidates, generations[1]));
// check overlap with L0 compacting sstables to make sure we are not generating overlap in L1.
Iterable<SSTableReader> compactingL0 = Iterables.filter(generations[0], Predicates.in(compacting));
if (!Sets.intersection(candidates, compacting).isEmpty() || !overlapping(candidates, compactingL0).isEmpty())
return Collections.emptyList();
}
return candidates.size() > 1 ? candidates : Collections.<SSTableReader>emptyList();
}
// for non-L0 compactions, pick up where we left off last time
Collections.sort(generations[level], SSTable.sstableComparator);
int start = 0; // handles case where the prior compaction touched the very last range
for (int i = 0; i < generations[level].size(); i++)
{
SSTableReader sstable = generations[level].get(i);
if (sstable.first.compareTo(lastCompactedKeys[level]) > 0)
{
start = i;
break;
}
}
// look for a non-suspect table to compact with, starting with where we left off last time,
// and wrapping back to the beginning of the generation if necessary
for (int i = 0; i < generations[level].size(); i++)
{
SSTableReader sstable = generations[level].get((start + i) % generations[level].size());
Set<SSTableReader> candidates = Sets.union(Collections.singleton(sstable), overlapping(sstable, generations[level + 1]));
if (Iterables.any(candidates, suspectP))
continue;
if (Sets.intersection(candidates, compacting).isEmpty())
return candidates;
}
// all the sstables were suspect or overlapped with something suspect
return Collections.emptyList();
}
private List<SSTableReader> ageSortedSSTables(Collection<SSTableReader> candidates)
{
List<SSTableReader> ageSortedCandidates = new ArrayList<SSTableReader>(candidates);
Collections.sort(ageSortedCandidates, SSTable.maxTimestampComparator);
return ageSortedCandidates;
}
public static File tryGetManifest(ColumnFamilyStore cfs)
{
return cfs.directories.tryGetLeveledManifest();
}
public synchronized void serialize()
{
File manifestFile = cfs.directories.getOrCreateLeveledManifest();
File oldFile = new File(manifestFile.getPath().replace(EXTENSION, "-old.json"));
File tmpFile = new File(manifestFile.getPath().replace(EXTENSION, "-tmp.json"));
JsonFactory f = new JsonFactory();
try
{
FileOutputStream fos = new FileOutputStream(tmpFile);
JsonGenerator g = f.createJsonGenerator(fos, JsonEncoding.UTF8);
g.useDefaultPrettyPrinter();
g.writeStartObject();
g.writeArrayFieldStart("generations");
for (int level = 0; level < generations.length; level++)
{
g.writeStartObject();
g.writeNumberField("generation", level);
g.writeArrayFieldStart("members");
for (SSTableReader ssTableReader : generations[level])
g.writeNumber(ssTableReader.descriptor.generation);
g.writeEndArray(); // members
g.writeEndObject(); // generation
}
g.writeEndArray(); // for field generations
g.writeEndObject(); // write global object
g.flush();
fos.getFD().sync();
g.close();
}
catch (IOException e)
{
throw new FSWriteError(e, tmpFile);
}
if (oldFile.exists() && manifestFile.exists())
FileUtils.deleteWithConfirm(oldFile);
if (manifestFile.exists())
FileUtils.renameWithConfirm(manifestFile, oldFile);
assert tmpFile.exists();
FileUtils.renameWithConfirm(tmpFile, manifestFile);
logger.debug("Saved manifest {}", manifestFile);
}
@Override
public String toString()
{
return "Manifest@" + hashCode();
}
public int getLevelCount()
{
for (int i = generations.length - 1; i >= 0; i--)
{
if (generations[i].size() > 0)
return i;
}
return 0;
}
public synchronized SortedSet<SSTableReader> getLevelSorted(int level, Comparator<SSTableReader> comparator)
{
return ImmutableSortedSet.copyOf(comparator, generations[level]);
}
public List<SSTableReader> getLevel(int i)
{
return generations[i];
}
public synchronized int getEstimatedTasks()
{
long tasks = 0;
long[] estimated = new long[generations.length];
for (int i = generations.length - 1; i >= 0; i--)
{
List<SSTableReader> sstables = generations[i];
estimated[i] = Math.max(0L, SSTableReader.getTotalBytes(sstables) - maxBytesForLevel(i)) / maxSSTableSizeInBytes;
tasks += estimated[i];
}
logger.debug("Estimating {} compactions to do for {}.{}",
new Object[] {Arrays.toString(estimated), cfs.table.name, cfs.columnFamily});
return Ints.checkedCast(tasks);
}
}