@cryptotaxi247 / netdata-1
Code Commits Issues Pulls Sessions CI/CD
master
master
md 172 lines 6.69 KB
Rendered Raw
1 <!--startmeta
2 custom_edit_url: "https://github.com/netdata/netdata/edit/master/src/collectors/ebpf.plugin/integrations/ebpf_sync.md"
3 meta_yaml: "https://github.com/netdata/netdata/edit/master/src/collectors/ebpf.plugin/metadata.yaml"
4 sidebar_label: "eBPF Sync"
5 learn_status: "Published"
6 learn_rel_path: "Collecting Metrics/Collectors/Storage and Filesystems"
7 keywords: ['syscall', 'eBPF', 'hard disk', 'memory']
8 message: "DO NOT EDIT THIS FILE DIRECTLY, IT IS GENERATED BY THE COLLECTOR'S metadata.yaml FILE"
9 endmeta-->
10
11 # eBPF Sync
12
13
14 <img src="https://netdata.cloud/img/ebpf.jpg" width="150"/>
15
16
17 Plugin: ebpf.plugin
18 Module: sync
19
20 <img src="https://img.shields.io/badge/maintained%20by-Netdata-%2300ab44" />
21
22 ## Overview
23
24 Monitor syscall responsible to move data from memory to storage device.
25
26 Attach tracing (kprobe, trampoline) to internal kernel functions according options used to compile kernel.
27
28 This collector is only supported on the following platforms:
29
30 - Linux
31
32 This collector supports collecting metrics from multiple instances of this integration, including remote instances.
33
34 The plugin needs setuid because it loads data inside kernel. Netada sets necessary permission during installation time.
35
36 ### Default Behavior
37
38 #### Auto-Detection
39
40 The plugin checks kernel compilation flags (CONFIG_KPROBES, CONFIG_BPF, CONFIG_BPF_SYSCALL, CONFIG_BPF_JIT, CONFIG_HAVE_SYSCALL_TRACEPOINTS), files inside debugfs, and presence of BTF files to decide which eBPF program will be attached.
41
42 #### Limits
43
44 The default configuration for this integration does not impose any limits on data collection.
45
46 #### Performance Impact
47
48 This thread will add overhead every time that an internal kernel function monitored by this thread is called. The estimated additional period of time is between 90-200ns per call on kernels that do not have BTF technology.
49
50 ## Setup
51
52
53 ### Prerequisites
54
55 #### Compile kernel
56
57 Check if your kernel was compiled with necessary options (CONFIG_KPROBES, CONFIG_BPF, CONFIG_BPF_SYSCALL, CONFIG_BPF_JIT) in `/proc/config.gz` or inside /boot/config file. Some cited names can be different accoring preferences of Linux distributions.
58 When you do not have options set, it is necessary to get the kernel source code from https://kernel.org or a kernel package from your distribution, this last is preferred. The kernel compilation has a well definedd pattern, but distributions can deliver their configuration files
59 with different names.
60
61 Now follow steps:
62 1. Copy the configuration file to /usr/src/linux/.config.
63 2. Select the necessary options: make oldconfig
64 3. Compile your kernel image: make bzImage
65 4. Compile your modules: make modules
66 5. Copy your new kernel image for boot loader directory
67 6. Install the new modules: make modules_install
68 7. Generate an initial ramdisk image (`initrd`) if it is necessary.
69 8. Update your boot loader
70
71
72 #### Debug Filesystem
73
74 This thread needs to attach a tracepoint to monitor when a process schedule an exit event. To allow this specific feaure, it is necessary to mount `debugfs` (`mount -t debugfs none /sys/kernel/debug`).
75
76
77
78 ### Configuration
79
80 #### Options
81
82 This configuration file have two different sections. The `[global]` overwrites all default options, while `[syscalls]` allow user to select the syscall to monitor.
83
84
85 <details open><summary>Config options</summary>
86
87
88
89 | Option | Description | Default | Required |
90 |:-----|:------------|:--------|:---------:|
91 | update every | Data collection frequency. | 10 | no |
92 | ebpf load mode | Define whether plugin will monitor the call (`entry`) for the functions or it will also monitor the return (`return`). | entry | no |
93 | apps | Enable or disable integration with apps.plugin | no | no |
94 | cgroups | Enable or disable integration with cgroup.plugin | no | no |
95 | pid table size | Number of elements stored inside hash tables used to monitor calls per PID. | 32768 | no |
96 | ebpf type format | Define the file type to load an eBPF program. Three options are available: `legacy` (Attach only `kprobe`), `co-re` (Plugin tries to use `trampoline` when available), and `auto` (plugin check OS configuration before to load). | auto | no |
97 | ebpf co-re tracing | Select the attach method used by plugin when `co-re` is defined in previous option. Two options are available: `trampoline` (Option with lowest overhead), and `probe` (the same of legacy code). | trampoline | no |
98 | maps per core | Define how plugin will load their hash maps. When enabled (`yes`) plugin will load one hash table per core, instead to have centralized information. | yes | no |
99 | lifetime | Set default lifetime for thread when enabled by cloud. | 300 | no |
100 | sync | Enable or disable monitoring for syscall `sync` | yes | no |
101 | msync | Enable or disable monitoring for syscall `msync` | yes | no |
102 | fsync | Enable or disable monitoring for syscall `fsync` | yes | no |
103 | fdatasync | Enable or disable monitoring for syscall `fdatasync` | yes | no |
104 | syncfs | Enable or disable monitoring for syscall `syncfs` | yes | no |
105 | sync_file_range | Enable or disable monitoring for syscall `sync_file_range` | yes | no |
106
107
108 </details>
109
110
111
112 #### via File
113
114 The configuration file name for this integration is `ebpf.d/sync.conf`.
115
116 The file format is a modified INI syntax. The general structure is:
117
118 ```ini
119 [section1]
120 option1 = some value
121 option2 = some other value
122
123 [section2]
124 option3 = some third value
125 ```
126 You can edit the configuration file using the [`edit-config`](https://github.com/netdata/netdata/blob/master/docs/netdata-agent/configuration/README.md#edit-configuration-files) script from the
127 Netdata [config directory](https://github.com/netdata/netdata/blob/master/docs/netdata-agent/configuration/README.md#locate-your-config-directory).
128
129 ```bash
130 cd /etc/netdata 2>/dev/null || cd /opt/netdata/etc/netdata
131 sudo ./edit-config ebpf.d/sync.conf
132 ```
133
134 ##### Examples
135 There are no configuration examples.
136
137
138
139 ## Alerts
140
141
142 The following alerts are available:
143
144 | Alert name | On metric | Description |
145 |:------------|:----------|:------------|
146 | [ sync_freq ](https://github.com/netdata/netdata/blob/master/src/health/health.d/synchronization.conf) | mem.sync | number of sync() system calls. Every call causes all pending modifications to filesystem metadata and cached file data to be written to the underlying filesystems. |
147
148
149 ## Metrics
150
151 Metrics grouped by *scope*.
152
153 The scope defines the instance that the metric belongs to. An instance is uniquely identified by a set of labels.
154
155
156
157 ### Per eBPF Sync instance
158
159 These metrics show total number of calls to functions inside kernel.
160
161 This scope has no labels.
162
163 Metrics:
164
165 | Metric | Dimensions | Unit |
166 |:------|:----------|:----|
167 | mem.file_sync | fsync, fdatasync | calls/s |
168 | mem.memory_map | msync | calls/s |
169 | mem.sync | sync, syncfs | calls/s |
170 | mem.file_segment | sync_file_range | calls/s |
171
172