TCP/IP网络互连 英文版 第3卷 客户/服务器编程及应用Linux/POSIX Sockets版PDF|Epub|txt|kindle电子书版本网盘下载

TCP/IP网络互连 英文版 第3卷 客户/服务器编程及应用Linux/POSIX Sockets版PDF格式电子书版下载

注意：本站所有压缩包均有解压码： 点击下载压缩包解压工具

1．1 Internet Applications Using TCP/IP1

1．2 Designing Applications For ADistribuled Environment1

Chapter 1 Introduction And Overview1

1．3 Standard And Nonstandard Application Protocols2

1．4 An Example Of Standard Application Protocol Use2

1．5 An Example TELNET Connection3

1．6 Using TELNET To Access An Alternative Service4

1．7 Application Protocols And Software Flexibility5

1．9 The Remainder Of This Text6

1．8 Viewing Services From The Provider s Perspective6

1．10 Summary7

Chapter 2 The Client Server Model And Software Design9

2．1 Introduction9

2．2 Motivation10

2．3 Terminology And Concepts10

2．3．1 Clients And Servers11

2．3．2 Privilege And Complexity11

2．3．4 Parameterization Of Clients12

2．3．3 Standard Vs. Nonstandard Client Software12

2．3．5 Connectionless Vs. Connection-Oriented Servers13

2．3．6 Stateless Vs. Stateful Servers14

2．3．7 A Stateless File Server Example15

2．3．8 A Stateful File Server Example15

2．3．9 Identifying A Client16

2．3．10 Statelessness Is A Protocol Issue18

2．3．11 Servers As Clients19

2．4 Summary20

3．2 Concurrency In Networks23

Chapter 3 Concurrent Processing In Client-Server Software23

3．1 Introduction23

3．3 Concurrency In Servers25

3．4 Terminology And Concepts26

3．4．1The Process Concept26

3．4．2 Sharing Of Local And Global Variables27

3．4．3 Procedure Calls28

3．5 An Example Of Concurrent Process Creation29

3．5．1 A Sequential C Example29

3．5．2 A Concurrent Version30

3．5．3 Timeslicing31

3．5．4 Singly-Threaded Process Assumption32

3．5．5 Making Processes Diverge33

3．6 Executing New Code34

3．7 Context Switching And Protocol Software Design34

3．8 Concurrency And Asynchronous I/O35

3．9 Summary36

4．1 Introduction39

4．2 Loosely Specified Protocol Software Interface39

Chapter 4 Application Interface To Protocols39

4．2．1 Advantages And Disadvantages40

4．3 Interface Functionality40

4．4 Conceptual Interface Specification41

4．5 System Calls42

4．6 Two Basic Approaches To Network Communication43

4．7 The Basic I/O Functions Available In Linux43

4．8 Using Linux I/O With TCP/IP45

4．9 Summary45

5．2 Berkeley Sockets47

Chapter 5 The Socket API47

5．1 Introduction47

5．3 Specifying A Protocol Interface48

5．4 The Socket Abstraction49

5．4．1 Socket Descriptors And File Descriptors49

5．4．2 System Data Structures For Sockets50

5．4．3 Making A Socket Active Or Passive51

5．5 Specifying An Endpoint Address52

5．6 A Generic Address Structure52

5．7．2 The Connect Call54

5．7 Major System Calls In The Socket API54

5．7．1 The Socket Call54

5．7．3 The Send Call55

5．7．4 The Recv Call55

5．7．5 The Close Call55

5．7．6 The Bind Call56

5．7．7 The Listen Call56

5．7．8 The Accept Call56

5．7．9 Using Read And Write With Sockets56

5．7．10 Summary Of Socket Calls57

5．8 Utility Routines For Integer Conversion58

5．9 Using Socket Calls In A Program58

5．10 Symbolic Constants For Socket Call Parameters59

5．11 Summary60

Chapter 6 Algorithms And Issues In Client Software Design63

6．1 Introduction63

6．2 Learning Algorithms Instead Of Details63

6．3 Client Architecture64

6．4 Identifying The Location Of A Server64

6．5 Parsing An Address Argument66

6．6 Looking Up A Domain Name67

6．7 Looking Up A Well-Known Port By Name68

6．8 Port Numbers And Network Byte Order68

6．9 Looking Up A Protocol By Name69

6．10 The TCP Client Algorithm69

6．11 Allocating A Socket70

6．12 Choosing A Local Protocol Port Number71

6．13 A Fundamental Problem In Choosing A Local IP Address71

6．15 Communicating With The Server Using TCP72

6．14 Connecting A TCP Socket To A Server72

6．16 Receiving A Response From A TCP Connectior73

6．17 Closing A TCP Connection74

6．17．1 The Need For Partial Close74

6．17．2 A Partial Close Operation74

6．18 Programming A UDP Client75

6．19 Connected And Unconnected UDP Sockets76

6．20 Using Connect With UDP76

6．21 Communicating With A Server Using UDP76

6．24 A Warning About UDP Unreuaouuy77

6．25 Summary77

6．22 Closing A Socket That Uses UDP77

6．23 Partial Close For UDP77

Chapter 7 Example Client Software81

7．1 Introduction81

7．2 The Importance Of Small Examples81

7．3 Hiding Details82

7．4 An Example Procedure Library For Client Programs82

7．5 Implementation Of ConnectTCP83

7．6 Implementation Of ConnectUDP84

7．7 A Procedure That Forms Connections85

7．8 Using The Example Library88

7．9 The DAYTIME Service88

7．10 Implementation Of A TCP Client For DAYTIME89

7．11 Reading From A TCP Connection90

7．12 The TIME Service91

7．13 Accessing The TIME Service91

7．14 Accurate Times And Network Delays92

7．15 A UDP Client For The TIME Service92

7．17 A TCP Client For The ECHO Service94

7．16 The ECHO Service94

7．18 A UDP Client For The ECHO Service96

7．19 Summary98

Chapter 8 Algorithms And Issues In Server Software Design101

8．1 Introduction101

8．2 The Conceptual Server Algorithm101

8．3 Concurrent Vs. Iterative Servers102

8．4 Connection-Oriented Vs. Connectionless Access102

8．5．2 UDP Semantics103

8．5 Transport Protocol Semantics103

8．5．1 TCP Semantics103

8．6 Choice Of Transport104

8．7 Connection-Oriented Servers104

8．8 Connectionless Servers105

8．9 Failure, Reliability, And Statelessness106

8．10 Optimizing Stateless Servers106

8．11 Four Basic Types Of Servers109

8．12 Request Processing Time109

8．14 An Iterative, Connection-Oriented Server Algorithm110

8．13 Iterative Server Algorithms110

8．15 Binding To A Well-Known Address Using INADDR_ANY111

8．16 Placing The Socket In Passive Mode112

8．17 Accepting Connections And Using Them112

8．18 An Iterative, Connectionless Server Algorithm112

8．19 Forming A Reply Address In A Connectionless Server113

8．20 Concurrent Server Algorithms114

8．21 Master And Slaves114

8．22 A Concurrent, Connectionless Server Algorithm115

8．23 A Concurrent, Connection-Oriented Server Algorithm116

8．24 Implementations Of Server Concurrency117

8．25 Using Separate Programs As Slaves118

8．26 Apparent Concurrency Using A Single Thread118

8．27 When To Use Each Server Type119

8．28 A Summary of Server Types120

8．29 The Important Problem Of Server Deadlock121

8．30 Alternative Implementations122

8．31 Summary122

9．2 Creating A Passive Socket125

Chapter 9 Iterative, Connectionless Servers (UDP)125

9．1 Introduction125

9．3 Process Structure129

9．4 An Example TIME Server130

9．5 Summary132

Chapter 10 Iterative, Connection-Oriented Servers (TCP)135

10．1 Introduction135

10．2 Allocating A Passive TCP Socket135

10．4 Process Structure136

10．3 A Server For The DAYTIME Service136

10．5 An Example DAYTIME Server137

10．6 Closing Connections140

10．7 Connection Termination And Server Vulnerability140

10．8 Summary141

Chapter 11 Concurrent, Connection-Oriented Servers (TCP)143

11．1 Introduction143

11．2 ECHO Service143

11．4 Process Structure144

11．3 Iterative Vs. Concurrent Implementations144

11．5 An Example Concurrent ECHO Server145

11．6 Cleaning Up Errant Processes149

11．7 Summary150

Chapter 12 Using Threads For Concurrency (TCP)151

12．1 Introduction151

12．2 Overview Of Linux Threads151

12．3 Advantages Of Threads152

12．4 Disadvantages Of Threads153

12．5 Descriptors, Delay, And Exit153

12．7．1 Mutex154

12．7 Thread Coordination And Synchronization154

12．6 Thread Exit154

12．7．2 Semaphore155

12．7．3 Condition Variable155

12．8 An Example Server Using Threads156

12．9 Monitor And Control160

12．10 Summary161

13．1 Introduction163

13．2 Data-driven Processing In A Server163

Chapter 13 Single-Inread, Concurrent Servers (TCP)163

13．3 Data-Driven Processing With A Single Thread164

13．4 Process Structure Of A Single-Thread Server165

13．5 An Example Single-Thread ECHO Server166

13．6 Summary168

Chapter 14 Multiprotocol Servers (TCP, UDP)171

14．1 Introduction171

14．2 The Motivation For Reducing The Number Of Servers171

14．3 Multiprotocol Server Design172

14．4 Process Structure172

14．5 An Example Multiprotocol DAYTIME Server173

14．6 The Concept Of Shared Code177

14．7 Concurrent Multiprotocol Servers177

14．8 Summary178

Chapter 15 Multiservice Servers (TCP, UDP)179

15．1 Introduction179

15．2 Consolidating Servers179

15．3 A Connectionless, Multiservice Server Design180

15．4 A Connection-Oriented, Multiservice Server Design181

15．6 A Single-Thread, Multiservice Server Implementation182

15．5 A Concurrent, Connection-Oriented, Multiservice Server182

15．7 Invoking Separate Programs From A Multiservice Server183

15．8 Multiservice, Multiprotocol Designs185

15．9 An Example Multiservice Server185

15．10 Static and Dynamic Server Configuration192

15．11 The Super Server, Inetd193

15．12 An Example Inetd Server195

15．13 A List Of Server Variations197

15．14 Summary197

16．2 Choosing Between An Iterative And A Concurrent Design199

16．1 Introduction199

Chapter 16 Uniform, Efficient Management Of Server Concurrency199

16．3 Level Of Concurrency200

16．4 Demand-Driven Concurrency201

16．5 The Cost Of Concurrency201

16．6 Overhead And Delay202

16．7 Small Delays Can Matter202

16．8 Slave Preallocation204

16．8．1 Preallocation In Linux204

16．8．2 Preallocation In A Connection-Oriented Server205

16．8．3 Mutex, File Locking and Concurrent Calls To Accept206

16．8．4 Preallocation In A Connectionless Server207

16．8．5 Preallocation, Bursty Traffic, And NFS208

16．8．6 Preallocation On A Multiprocessor208

16．9 Delayed Slave Allocation208

16．10 The Uniform Basis For Both Techniques210

16．11 Combining Techniques210

16．12 Summary211

17．1 Introduction213

17．2 The Advantages Of Concurrency213

Chapter 17 Concurrency In Clients213

17．3 The Motivation For Exercising Control214

17．4 Concurrent Contact With Multiple Servers215

17．5 Implementing Concurrent Clients216

17．6 Single-Thread Implementations217

17．7 An Example Concurrent Client That Uses ECHO218

17．8 Execution Of The Concurrent Client222

17．9 Concurrency In The Example Code224

17．10 Summary224

18．2 Multiprotocol Environments227

Chapter 18 Tunneling At The Transport And Application Levels227

18．1 Introduction227

18．3 Mixing Network Techhotogtes229

18．4 Dynamic Circuit Allocation230

18．5 Encapsulation And Tunneling231

18．6 Tunneling Through An IP Internet232

18．7 Application-Level Tunneling Between Clients And Servers232

18．8 Tunneling, Encapsulation, And Dialup Phone Lines233

18．9 Summary234

19．2．1 The Reality Of Limited Access237

19．2 Clients And Servers In Constrained Environments237

Chapter 19 Application Level Gateways237

19．1 Introduction237

19．2．2 Computers With Limited Functionality238

19．2．3 Connectivity Constraints That Arise From Security238

19．3 Using Application Gateways238

19．4 Interoperability Through A Mail Gateway240

19．5 Implementation Of A Mail Gateway240

19．6 A Comparison Of Application Gateways And Tunneling241

19．7 Application Gateways And Limited Internet Connectivity243

19．9 Application Gateways And The Extra Hop Problem245

19．8 Application Gateways Used For Sequrity245

19．10 An Example Application Gateway247

19．11 Implementation Of An Application Gateway248

19．12 Code For The Application Gateway250

19．13 An Example Gateway Exchange251

19．14 Using Rfcd With forward Or Slocal252

19．15 A General-Purpose Application Gateway253

19．16 Operation Of SLIRP253

19．18 IP Addressing And SLIRP254

19．17 How SLIRP Handles Connections254

19．19 Summary255

Chapter 20 External Data Representation (XDR)257

20．1 Introduction257

20．2 Representations Of Data257

20．3 Asymmetric Conversion And The N-Squared Problem258

20．4 Network Standard Byte Order259

20．5 A De Facto Standard External Data Representation261

20．6 XDR Data Types261

20．9 XDR Library Routines263

20．7 Implicit Types263

20．8 Software Support For Using XDR263

20．10 Building A Message One Piece At A Time264

20．11 Conversion Routines In The XDR Library265

20．12 XDR Streams, I/O, and TCP267

20．13 Records, Record Boundaries, And Datagram I/O268

20．14 Summary268

21．1 Introduction271

21．2 Remote Procedure Call Model271

Chapter 21 Remote Procedure Call Concept (RPC)271

21．3 Two Paradigms For Building Distributed Programs272

21．4 A Conceptual Model For Conventional Prccedure Calls273

21．5 An Extension Of the Procedural Model274

21．6 Execution Of Conventional Procedure Call And Return274

21．7 The Procedural Model In Distributed System275

21．8 Analogy Between Client-Server And RPC276

21．9 Distributed Computation As A Program277

21．11 Remote Programs And Procedures278

21．10 Sun Microsystems Remote Procedure Call Definition278

21．12 Reducing The Number Of Arguments279

21．13 Identifying Remote Programs And Procedures279

21．14 Accommodating Multiple Versions Of A Remote Program280

21．15 Mutual Exclusion For Procedures In A Remote Program281

21．16 Communication Semantics282

21．17 At Least Once Semantics282

21．18 RPC Retransmission283

21．19 Mapping A Remote Program To A Protocol Port283

21．20 Dynamic Port Mapping284

21．21 RPC Port Mapper Algorithm285

21．22 ONC RPC Message Format287

21．23 Marshaling Arguments For A Remote Procedure288

21．24 Authentication288

21．25 An Example Of RPC Message Representation290

21．26 An Example Of The UNIX Authentication Field290

21．27 Summary291

Chapter 22 Distributed Program Generation (Rpegen Concept)295

22．1 Introduction295

22．2 Using Remote Procedure Calls296

22．3 Programming Mechanisms To Support RPC297

22．4 Dividing A Program Into Local And Remote Procedures298

22．5 Adding Code For RPC299

22．6 Stub Procedures299

22．7 Multiple Remote Procedures And Dispatching300

22．8 Name Of The Client-Side Stub Procedure301

22．9 Using Rpcgen To Generate Distributed Programs302

22．10 Rpcgen Output And Interface Procedures302

22．12 Using Rpcgen To Build A Client And Server304

22．11 Rpcgen Input And Output304

22．13 Summary305

Chapter 23 Distributed Program Generation (Rpcgen Example)307

23．1 Introduction307

23．2 An Example To Illustrate Rpcgen308

23．3 Dictionary Operations308

23．4 Eigth Steps To A Distributed Application309

23．5 Step 1: Build A Conventional Applicaton Program310

23．6 Step 2: Divide The Program Into Two Parts314

23．7 Step 3: Create An Rpcgen Specification320

23．8 Step 4: Run Rpcgen322

23．9 The .h File Produced By Rpcgen322

23．10 The XDR Conversion File Produced By Rpcgen325

23．11 The Client Code Produced By Rpcgen325

23．12 The Server Code Produced By Rpcgen328

23．13 Step 5: Write Stub Interface Procedures331

23．13．1 Client-Side Interface Routines331

23．13．2 Server-Side Interface Routines333

23．14 Step 6: Compile And Link The Client Program335

23．15 Step 7: Compile And Link The Server Program339

23．17 Using The Make Utility341

23．16 Step 8 : Start The Server And Execute The Client341

23．18 Summary344

Chapter 24 Network File System Concepts (NFS)347

24．1 Introduction347

24．2 Remote File Access Vs. Transfer347

24．3 Operations On Remote Files348

24．4 File Access Among Heterogeneous Computers348

24．7 Review Of The UNIX File System349

24．5 Stateless Servers349

24．6 NFS And UNIX File Semantics349

24．7．1 Basic Defintions350

24．7．2 A Byte Sequence Without Record Boundaries350

24．7．3 A File s Owner And Group Identifiers350

24．7．4 Protection And Access351

24．7．5 The Open-Read-Write-Close Paradigm352

24．7．6 Data Transfer353

24．7．7 Permission To Search A Directory353

24．7．9 Seeking Beyond The End Of File354

24．7．8 Random Access354

24．7．10 File Position And Concurrent Access355

24．7．11 Semantics Of Write During Concurrent Access*356

24．7．12 File Names And Paths356

24．7．13 Inode: Information Stored With A File357

24．7．14 Stat Operation358

24．7．15 The File Naming Mechanism359

24．7．16 File System Mounts360

24．7．17 File Name Resolution362

24．7．18 Symbolic Links363

24．8 Files Under NFS363

24．9 NFS File Types364

24．10 NFS File Modes364

24．11 NFS File Attributes365

24．12 NFS Client And Server366

24．13 NFS Client Operation367

24．14 NFS Client And UNIX Systems368

24．15 NFS Mounts369

24．17 Handles Replace Path Names370

24．16 File Handle370

24．18 File Positioning With A Stateless Server372

24．19 Operations On Directories372

24．20 Reading A Directory Statelessly372

24．21 Multiple Hierarchies In An NFS Server373

24．22 The Mount Protocol374

24．23 Transport Protocols For NFS374

24．24 Summary375

25．2 Using RPC To Define A Protocol377

25．1 Introduction377

Chapter 25 Network File System Protocols (NFS, Mount)377

25．3 Defining A Protocol With Data Structures And Procedures378

25．4 NFS Constant, Type, And Data Declarations379

25．4．1 NFS Constants379

25．4．2 NFS Typedef Declarations380

25．4．3 NFS Data Structures380

25．5 NFS Procedures383

25．6．6 NFSPROC3_READLINK (Procedure 5)385

25．6．5 NFSPROC3_ACCESS (Procedure 4)385

25．6．4 NFSPROC3_LOOKVP (Procedure 3)385

25．6．2 NFSPROC3_GETATTR (Procedure 1)385

25．6．1 NFSPROC3_NULL (Procedure 0)385

25．6 Semantics Of NFS Operations385

25．6．3 NFSPROC3_SETATTR (Procedure 2)385

25．6．7 NFSPROC3_READ (Procedure 6)386

25．6．8 NFSPROC3_WRITE (Procedure 7)386

25．6．9 NFSPROC3_CREATE (Procedure 8)386

25．6．10 NFSPROC3_MKDIR (Procedure 9)386

25．6．11 NFSPROC3_SYMLINK (Procedure 10)386

25．6．12 NFSPROC3_ MKNOD (Procedure 11)386

25．6．16 NFSPROC3_LINK (Procedure 15)387

25．6．17 NFSPROC3_READDIR (Procedure 16)387

25．6．13 NFSPROC3_REMOVE (Procedure 12)387

25．6．15 NFSPROC3_RENAME (Procedure 14)387

25．6．14 NFSPROC3_RMDIR (Procedure 13)387

25．6．18 NFSPROC3_READDIRPLUS (Procedure 17)388

25．6．19 NFSPROC3_FSSTAT (Procedure 18)388

25．6．20 NFSPROC3_FSINFO (Procedure 19)388

25．6．21 NFSPROC3_PATHCONF (Procedure 20)388

25．7．2 Mount Type Definitions389

25．7．1 Mount Constant Definitions389

25．6．22 NFSPROC3_ COMMIT (Procedure 21)389

25．7 The Mount Protocol389

25．7．3 Mount Data Structures390

25．8 Procedures In The Mount Protocol391

25．9 Semantics Of Mount Operations391

25．9．4 MOUNTPROC3_UMNT(Procedure 3)392

25．9．6 MOUNTPROC3_EXPORT(Procedure 5)392

25．9．5 MOUNTPROC3_UMNTALL(Procedure 4)392

25．9．3 MOUNTPROC3_DUMP(Procedure 2)392

25．9．2 MOUNTPROC3_MNT(Procedure 1)392

25．9．1 MOUNTPROC3_NULL(Procedure 0)392

25．10 NFS And Mount Authentication393

25．11 File Locking394

25．12 Changes In NFS Between Versions 3 And 4394

25．13 Summary395

Chapter 26 A TELNET Client (Program Structure)397

26．1 Introduction397

26．2．3 Terminals, Windows, and Files398

26．2．2 Command And Control Information398

26．2．1 The User s Terminal398

26．2 Overview398

26．2．4 The Need For Concurrency399

26．2．5 A Process Model For A TELNET Client400

26．3 A TELNET Client Algorithm400

26．4 Terminal I/O In Linux401

26．4．1 Controlling A Device Driver402

26．5 Establishing Terminal Modes403

26．6 Global Variable Used For Stored State405

26．7 Restoring Terminal Modes Before Exit406

26．8 Client Suspension And Resumption407

26．9 Finite State Machine Specification408

26．10 Embedding Commands In A TELNET Data Stream409

26．11 Option Negotiation410

26．12 Request/Offer Symmetry410

26．13 TELNET Character Definitions411

26．14 A Finite State Machine For Data From The Server412

26．15 Transitions Among States413

26．17 A Compact FSM Representation415

26．16 A Finite State Machine Implementation415

26．18 Keeping The Compact Representation At Run-Time417

26．19 Implementation Of A Compact Representation417

26．20 Building An FSM Transition Matrix419

26．21 The Socket Output Finite State Machine421

26．22 Definitions For The Socket Output FSM423

26．23 The Option Subnegotiation Finite State Machine424

26．24 Definitions For The Option Subnegotiation FSM425

26．25 FSM Initialization426

26．26 Arguments For The TELNET Client427

26．27 The Heart Of The TELNET Client428

26．28 Implementation Of The Main FSM432

26．29 Summary433

Chapter 27 A TELNET Client (Implementation Details)435

27．1 Introduction435

27．2 The FSM Action Procedures435

27．3 Recording The Type Of An Option Request436

27．4 Performing No Operation437

27．5 Responding To WILL/WONT For The Echo Option437

27．7 Responding To WILL/WONT For The No Go-Ahead Option439

27．6 Responding To WILL/WONT For Unsupported Options439

27．8 Generating DO/DONT For Binary Transmission441

27．9 Responding To DO/DONT For Unsupported Options442

27．10 Responding To DO/DONT For Transmit Binary Option442

27．11 Responding To DO/DONT For The Terminal Type Option444

27．12 Option Subnegotiation445

27．13 Sending Terminal Type Information446

27．14 Terminating Subnegotiation448

27．15 Sending A Character To The Server448

27．16 Displaying Incoming Data On The User s Terminal450

27．17 Using Termcap To Control The User s Terminal453

27．18 Writing A Block Of Data To The Server455

27．19 Interacting With The Client Process456

27．20 Responding To Illegal Commands456

27．21 Scripting To A File457

27．22 Implementation Of Scripting457

27．23 Initialization Of Scripting458

27．24 Collecting Characters Of The Script File Name459

27．25 Opening A Script File460

27．26 Terminating Scripting462

27．27 Printing Status Information463

27．28 Summary464

Chapter 28 Streaming Audio And Video Transport (RTP Concept And Design)467

28．1 Introduction467

28．2 Streaming Service467

28．3 Real-Time Delivery468

28．4 Protocol Compensation For Jitter468

28．5 Retransmission, Loss, And Recovery469

28．6 Real-Time Transport Protocol470

28．8 Delayed Playback And Jitter Buffers471

28．7 Stream Translation And Mixing471

28．9 RTP Control Protocol (RTCP)472

28．10 Synchronizing Multiple Sireams474

28．11 RTP Transport And Many-To-Many Transmission475

28．12 Sessions, Streams, Protocol Ports, And Demultiplexing476

28．13 Basic Approaches To Encoding477

28．14 Conceptual Organization Of RTP Software478

28．15 Process/Thread Structure479

28．16 Semantics Of The API481

28．17 Jitter Buffer Design And Rebuffering482

28．18 Event Handling483

28．19 Playback Anomaly And Timestamp Complications483

28．20 Size Of An Example Real-Time Library484

28．21 An Example MP3 Player484

28．22 Summary485

Chapter 29 Streaming Audio And Video Transport (Example RTP Implementation)487

29．1 Introduction487

29．2 An Integrated Implementation487

29．4 RTP Definitions488

29．3 Program Architecture488

29．5 Manipulation Of Time Values492

29．6 RTP Sequence Space Processing493

29．7 RTP Packet Queue Manipulation494

29．8 RTP Input Processing496

29．9 Keeping Statistics For RTCP499

29．10 RTP Initialization500

29．11 RTCP Definitions504

29．12 Receiving RTCP Sender Reports506

29．13 Generating RTCP Receiver Reports507

29．14 RTCP Header Creation509

29．15 RTCP Delay Computation510

29．16 Coneration Of An RTCP Bye Message511

29．17 Size Of An Integrated Implementation511

29．18 Summary512

Chapter 30 Practical Hints And Techniques For Linux Servers515

30．1 Introduction515

30．2 Operating In Background515

30．3 Programming A Server To Operate In Background516

30．4 Open Descriptors And Inheritance517

30．6 Signals From The Controlling TTY518

30．5 Programming A Server To Close Inherited Descriptors518

30．7 Programming A Server To Change Its Controlling TTY519

30．8 Moving To A Safe And Known Directory519

30．9 Programming A Server To Change Directories520

30．10 The Linux Umask520

30．11 Programming A Server To Set Its Umask521

30．12 Process Groups521

30．13 Programming A Server To Set Its Process Group521

30．16 Mutual Exclusion For The Server522

30．14 Descriptors For Standard I/O522

30．15 Programming A Server To Open Standard Descriptors522

30．17 Programming A Server To Avoid Multiple Copies523

30．18 Recording A Server s Process ID524

30．19 Programming A Server To Record Its Process ID524

30．20 Waiting For A Child Process To Exit524

30．23 Programming A Server To Ignore Extraneous Signals525

30．24．1 Generating Log Messages525

30．24 Using A System Log Facility525

30．22 Extraneous Signals525

30．21 Programming A Server To Wait For Each Child To Exit525

30．24．2 The Advantage Of Indirection And Standard Error526

30．24．3 Limitations Of I/O Redirection527

30．24．4 A Client-Server Solution527

30．24．5 The Syslog Mechanism528

30．24．6 Syslog Message Classes528

30．24．7 Syslog Facilities528

30．24．8 Syslog Priority Levels529

30．24．9 Using Syslog529

30．24．10 An Example Syslog Configuration File530

30．25 Sumary531

Chapter 31 Deadlock And Starvation In Client-Server Systems533

31．1 Introduction533

31．2 Definition Of Deadlock534

31．3 Difficulty Of Deadlock Detection534

31．4 Deadlock Avoidance535

31．5 Deadlock Between A Client And Server535

31．6 Avoiding Deadlock In A Single Interaction536

31．7 Starvation Among A Set Of Clients And A Server536

31．8 Busy Connections And Starvation537

31．9 Avoiding Blocking Operations538

31．10 Processes, Connections, And Other Limits538

31．11 Cycles Of Clients And Servers539

31．12 Documenting Dependencies540

31．13 Summary540

Appendix 1 System Calls And Library Routines Used With Sockets543

Appendix 2 Manipulation Of Linux File And Socket Descriptors573

Bibliography577

Index585