-
Notifications
You must be signed in to change notification settings - Fork 2.3k
/
SegmentFile.java
265 lines (233 loc) · 8.1 KB
/
SegmentFile.java
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
/*
* Copyright (c) 2002-2016 "Neo Technology,"
* Network Engine for Objects in Lund AB [http://neotechnology.com]
*
* This file is part of Neo4j.
*
* Neo4j is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as
* published by the Free Software Foundation, either version 3 of the
* License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
package org.neo4j.coreedge.raft.log.segmented;
import java.io.File;
import java.io.IOException;
import org.neo4j.coreedge.raft.log.EntryRecord;
import org.neo4j.coreedge.raft.log.EntryRecordCursor;
import org.neo4j.coreedge.raft.log.LogPosition;
import org.neo4j.coreedge.raft.log.RaftLogEntry;
import org.neo4j.coreedge.raft.replication.ReplicatedContent;
import org.neo4j.coreedge.raft.state.ChannelMarshal;
import org.neo4j.cursor.IOCursor;
import org.neo4j.io.fs.FileSystemAbstraction;
import org.neo4j.io.fs.StoreChannel;
import org.neo4j.kernel.impl.transaction.log.PhysicalFlushableChannel;
import org.neo4j.kernel.impl.transaction.log.ReadAheadChannel;
import org.neo4j.logging.Log;
import org.neo4j.logging.LogProvider;
import org.neo4j.storageengine.api.ReadPastEndException;
import org.neo4j.storageengine.api.WritableChannel;
import static org.neo4j.coreedge.raft.log.EntryRecord.read;
/**
* Keeps track of a segment of the RAFT log, i.e. a consecutive set of entries.
* A segment can have several concurrent readers but just a single writer.
*
* The single writer should perform all write and control operations,
* since these are not thread-safe.
*
* Concurrent reading is thread-safe.
*/
class SegmentFile
{
private static final SegmentHeader.Marshal headerMarshal = new SegmentHeader.Marshal();
private final Log log;
private final FileSystemAbstraction fileSystem;
private final File file;
private final StoreChannelPool readerPool;
private final ChannelMarshal<ReplicatedContent> contentMarshal;
private final SegmentHeader header;
private PhysicalFlushableChannel bufferedWriter;
private boolean markedForDisposal;
private Runnable onDisposal;
private volatile boolean isDisposed;
SegmentFile(
FileSystemAbstraction fileSystem,
File file,
ChannelMarshal<ReplicatedContent> contentMarshal,
LogProvider logProvider,
SegmentHeader header )
{
this.fileSystem = fileSystem;
this.file = file;
this.contentMarshal = contentMarshal;
this.header = header;
log = logProvider.getLog( getClass() );
readerPool = new StoreChannelPool( fileSystem, file, "r", logProvider );
}
static SegmentFile create(
FileSystemAbstraction fileSystem,
File file,
ChannelMarshal<ReplicatedContent> contentMarshal,
LogProvider logProvider,
SegmentHeader header ) throws IOException
{
SegmentFile segment = new SegmentFile( fileSystem, file, contentMarshal, logProvider, header );
if ( fileSystem.fileExists( file ) )
{
throw new IllegalStateException( "File was not expected to exist" );
}
headerMarshal.marshal( header, segment.getOrCreateWriter() );
return segment;
}
/**
* Channels must be closed when no longer used, so that they are released back to the pool of readers.
*/
IOCursor<EntryRecord> getReader( long logIndex ) throws IOException, DisposedException
{
assert logIndex > header.prevIndex();
long offsetIndex = logIndex - (header.prevIndex() + 1);
LogPosition position = findCachedStartingPosition( offsetIndex );
ReadAheadChannel<StoreChannel> reader =
new ReadAheadChannel<StoreChannel>( readerPool.acquire( position.byteOffset ) )
{
@Override
public void close()
{
// TODO: Consider including position of channel.
readerPool.release( channel );
}
};
long currentIndex = position.logIndex;
assert currentIndex == 0; // until we properly implement caching
try
{
while ( currentIndex < offsetIndex )
{
read( reader, contentMarshal );
currentIndex++;
}
}
catch ( ReadPastEndException e )
{
reader.close();
return IOCursor.getEmpty();
}
return new EntryRecordCursor( reader, contentMarshal, currentIndex, logPosition -> {
try
{
// TODO: Cache the end position.
// TODO: explain why we need the end position in the cache
reader.close();
}
catch ( IOException e )
{
log.error( "Failed to close reader: " + file );
}
} );
}
private LogPosition findCachedStartingPosition( long offsetIndex )
{
// TODO: Implement cache lookups.
return new LogPosition( 0, SegmentHeader.SIZE );
}
private PhysicalFlushableChannel getOrCreateWriter() throws IOException
{
if ( bufferedWriter == null )
{
StoreChannel channel = fileSystem.open( file, "rw" );
channel.position( channel.size() );
bufferedWriter = new PhysicalFlushableChannel( channel );
}
return bufferedWriter;
}
/**
* There is just a single writer and it is closed when the segment is disposed of.
*/
WritableChannel writer() throws IOException, DisposedException
{
if( markedForDisposal )
{
throw new DisposedException();
}
return getOrCreateWriter();
}
long position() throws IOException
{
return getOrCreateWriter().position();
}
void closeWriter()
{
if ( bufferedWriter != null )
{
try
{
flush();
bufferedWriter.close();
bufferedWriter = null;
checkFullDisposal();
}
catch ( IOException e )
{
log.error( "Failed to close writer: " + bufferedWriter, e );
}
}
}
public void write( long logIndex, RaftLogEntry entry ) throws IOException
{
EntryRecord.write( getOrCreateWriter(), contentMarshal, logIndex, entry.term(), entry.content() );
flush();
}
void flush() throws IOException
{
bufferedWriter.prepareForFlush().flush();
}
/**
* Marks this segment for eventual disposal. The segment will not be disposed of until
* all readers and writers to the segment file are closed.
*
* @param onDisposal Called when the segment is fully disposed with no readers and a closed writer.
* @throws DisposedException Thrown if this segment already is marked for disposal.
*/
void markForDisposal( Runnable onDisposal ) throws DisposedException
{
if( markedForDisposal )
{
throw new DisposedException();
}
this.onDisposal = onDisposal;
this.markedForDisposal = true;
readerPool.markForDisposal( this::checkFullDisposal );
}
private synchronized void checkFullDisposal()
{
if ( bufferedWriter == null && readerPool.isDisposed() )
{
isDisposed = true;
onDisposal.run();
}
}
boolean isDisposed()
{
return isDisposed;
}
public void delete()
{
fileSystem.deleteFile( file );
}
public SegmentHeader header()
{
return header;
}
public long size()
{
return fileSystem.getFileSize( file );
}
}