Red Hat ENTERPRISE LINUX 5.4 - SYSTEMTAP BEGINNERS GUIDE Informazioni Techniche Pagina 26
- Pagina / 240
- Indice
- RISOLUZIONE PROBLEMI
- SEGNALIBRI
Valutato. / 5. Basato su recensioni clienti
12 System Analysis and Tuning Guide
in
Interrupts per second. A high value indicates a high I/O level (network and/or
disk).
cs
Number of context switches per second. Simplified this means that the kernel has
to replace executable code of one program in memory with that of another pro-
gram.
us
Percentage of CPU usage from user processes.
sy
Percentage of CPU usage from system processes.
id
Percentage of CPU time spent idling. If this value is zero over a longer period of
time, your CPU(s) are working to full capacity. This is not necessarily a bad sign
—rather refer to the values in columns r and b to determine if your machine is
equipped with sufficient CPU power.
wa
If "wa" time is non-zero, it indicates throughput lost due to waiting for I/O. This
may be inevitable, for example, if a file is being read for the first time, back-
ground writeback cannot keep up, and so on. It can also be an indicator for a
hardware bottleneck (network or hard disk). Lastly, it can indicate a potential
for tuning the virtual memory manager (refer to Chapter15, Tuning the Memory
Management Subsystem (page179)).
st
Percentage of CPU time used by virtual machines.
See vmstat --help for more options.
2.1.2 System Activity Information: sar and
sadc
sar can generate extensive reports on almost all important system activities, among
them CPU, memory, IRQ usage, IO, or networking. It can either generate reports on
- SUSE Linux 1
- Enterprise Server 1
- Contents 3
- III Kernel Monitoring 69 4
- IV Resource Management 113 5
- V Kernel Tuning 153 6
- 16 Tuning the Network 187 7
- 17 Tracing Tools 193 7
- 18 kexec and kdump 205 7
- A GNU Licenses 221 8
- About This Guide 9
- 1 Available Documentation 10
- 2 Feedback 12
- 3 Documentation Conventions 13
- PartI.Basics 15
- General Notes on System 17
- 1.2 Rule Out Common Problems 18
- 1.3 Finding the Bottleneck 18
- 1.4 Step-by-step Tuning 19
- PartII.System Monitoring 21
- System Monitoring Utilities 23
- 2.1.1 vmstat 24
- With sadc 27
- CPU Utilization Report: sar 28
- Memory Usage Report: sar -r 28
- 2.1.2.3 Visualizing sar Data 30
- 2.2 System Information 31
- Viewer: udevadm monitor 34
- 2.3 Processes 35
- 2.3.2 Process List: ps 36
- 2.3.3 Process Tree: pstree 37
- 2.3.4 Table of Processes: top 38
- 2.4 Memory 42
- HugePages_Rsvd: 0 44
- HugePages_Surp: 0 44
- Hugepagesize: 2048 kB 44
- DirectMap4k: 2689024 kB 44
- DirectMap2M: 5691392 kB 44
- 2.5 Networking 45
- 2.6 The /proc File System 49
- 2.6.1 procinfo 51
- 2.7 Hardware Information 53
- 2.8 Files and File Systems 54
- Binaries 56
- 2.8.4 File Properties: stat 56
- 2.9 User Information 57
- 2.10 Time and Date 57
- 2.11 Graph Your Data: RRDtool 58
- 2.11.2.1 Collecting Data 59
- 2.11.2.2 Creating Database 60
- 2.11.3 For More Information 63
- Monitoring with Nagios 65
- 3.3.1 Object Definition Files 67
- 3.4 Configuring Nagios 69
- 3.5 Troubleshooting 72
- 3.6 For More Information 73
- Analyzing and Managing 75
- System Log Files 75
- 4.3 Managing Log Files with 78
- 4.4 Monitoring Log Files with 80
- Log Entries 81
- PartIII.Kernel Monitoring 83
- SystemTap—Filtering and 85
- Analyzing System Data 85
- 5.1.1 SystemTap Scripts 86
- 5.1.2 Tapsets 86
- 5.1.3 Commands and Privileges 87
- 5.2 Installation and Setup 88
- 5.3 Script Syntax 90
- Ctrl + C 91
- 5.3.3.1 Functions 92
- Variables 95
- Conditional Statements 96
- 5.4 Example Script 97
- 5.5 User-Space Probing 98
- 5.6 For More Information 99
- Kernel Probes 101
- 6.1 Supported Architectures 102
- 6.2 Types of Kernel Probes 102
- 6.3 Kernel probes API 103
- 6.4 Debugfs Interface 104
- 6.5 For More Information 105
- Perfmon2—Hardware 107
- Based Performance 107
- Monitoring 107
- 7.1.2 Sampling and Counting 108
- 7.2 Installation 109
- 7.3 Using Perfmon 110
- 7.4 Retrieving Metrics From 113
- interrupt = 115
- ovfl_intr_ns / 115
- ovfl_intr_all_count 115
- 7.5 For More Information 116
- OProfile—System-Wide 119
- Profiler 119
- 8.4 Using OProfile 120
- 8.4.1 General Steps 121
- 8.5 Using OProfile's GUI 123
- 8.6 Generating Reports 124
- 8.7 For More Information 124
- PartIV.Resource 127
- Management 127
- General System Resource 129
- 9.1.2 Installation Scope 130
- 9.1.3 Default Runlevel 130
- 9.2 Disabling Unnecessary 131
- Services 131
- 9.3.1 File Systems 132
- Kernel Control Groups 135
- 10.2 Scenario 136
- Disk I/O 137
- Network I/O 137
- 10.3 Control Group Subsystems 138
- 10.4 Using Controller Groups 141
- 10.4.3 Example: Cpusets 142
- 10.4.4 Example: cgroups 143
- 10.5 For More Information 144
- Power Management 147
- 11.1.4 Turbo Features 150
- 11.2 The Linux Kernel CPUfreq 151
- Infrastructure 151
- 11.3 Viewing, Monitoring and 153
- Tuning Power-related Settings 153
- Statistics with cpupower 157
- 11.4 Special Tuning Options 160
- 11.5 Creating and Using Power 163
- Management Profiles 163
- 11.6 Troubleshooting 164
- 11.7 For More Information 165
- PartV.Kernel Tuning 167
- Installing Multiple Kernel 169
- Versions 169
- 12.1 Enabling and Configuring 170
- Multiversion Support 170
- Kernel Versions with YaST 171
- Kernel Versions with zypper 172
- Tuning I/O Performance 175
- 13.2 Available I/O Elevators 176
- 13.2.2 NOOP 177
- 13.2.3 DEADLINE 177
- 13.3 I/O Barrier Tuning 178
- Tuning the Task Scheduler 179
- 14.2 Process Classification 180
- 14.3 O(1) Scheduler 181
- 14.4.1 How CFS Works 183
- 14.4.2 Grouping Processes 183
- 14.4.4 Terminology 184
- Scheduler Statistics 189
- 14.5 For More Information 190
- Tuning the Memory 193
- Management Subsystem 193
- 15.1 Memory Usage 194
- 15.1.4 Buffer Heads 195
- 15.1.5 Writeback 195
- 15.1.6 Readahead 195
- 15.1.7 VFS caches 195
- 15.2 Reducing Memory Usage 196
- Tunable Parameters 197
- 15.3.2 Writeback Parameters 198
- 15.3.3 Readahead parameters 199
- 15.3.4 Further VM Parameters 199
- 15.5 Monitoring VM Behavior 200
- Tuning the Network 201
- /proc/sys/net/ipv6 202
- /proc/sys/net/core 202
- /proc/sys/net/ipv4 202
- 16.3 Netfilter 203
- 16.4 For More Information 204
- PartVI.Handling 205
- System Dumps 205
- Tracing Tools 207
- Valgrind 212
- 17.3.3 General Information 213
- 17.3.4 Default Options 214
- 17.3.5 How Valgrind Works 214
- 17.3.6 Messages 215
- 17.3.7 Error Messages 216
- 17.4 For More Information 217
- 18.1 Introduction 219
- 18.2 Required Packages 220
- 18.3 kexec Internals 220
- 18.4 Basic kexec Usage 221
- Routine Reboots 222
- /sbin/mkinitrd -D vlan0 223
- /sbin/mkinitrd -I eth0 223
- 18.6.2 YaST Configuration 226
- 18.7 Analyzing the Crash Dump 227
- 18.7.1 Kernel Binary Formats 228
- 18.8 Advanced kdump 231
- Configuration 231
- 18.9 For More Information 232
- GNU Licenses 235
- 2. VERBATIM COPYING 236
- 3. COPYING IN QUANTITY 236
- 4. MODIFICATIONS 236
- 5. COMBINING DOCUMENTS 237
- GNU Licenses 225 239
Prodotti e manuali riguardandi Software Red Hat ENTERPRISE LINUX 5.4 - SYSTEMTAP BEGINNERS GUIDE
(75 pagine)© 2020, manymanuals.it. Tutti i diritti riservati | 0.044 s |
Manymanuals.com
Manymanuals.de
Manymanuals.fr
Manymanuals.it
Manymanuals.pl
Manymanuals.cz
Manymanuals.es
Manymanuals-pt.com
Commenti su questo manuale