blob: 320b1fc9f3a61519599fa82cc50c3801a8bbf008 (
plain)
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
|
<!--
title: "slabinfo.plugin"
custom_edit_url: "https://github.com/netdata/netdata/edit/master/collectors/slabinfo.plugin/README.md"
sidebar_label: "slabinfo.plugin"
learn_status: "Published"
learn_topic_type: "References"
learn_rel_path: "References/Collectors references/System metrics"
-->
# slabinfo.plugin
SLAB is a cache mechanism used by the Kernel to avoid fragmentation.
Each internal structure (process, file descriptor, inode...) is stored within a SLAB.
## configuring Netdata for slabinfo
The plugin is disabled by default because it collects and displays a huge amount of metrics.
To enable it set `slabinfo = yes` in the `plugins` section of the `netdata.conf` configuration file.
There is currently no configuration needed for the plugin itself.
As `/proc/slabinfo` is only readable by root, this plugin is setuid root.
## For what use
This slabinfo details allows to have clues on actions done on your system.
In the following screenshot, you can clearly see a `find` done on a ext4 filesystem (the number of `ext4_inode_cache` & `dentry` are rising fast), and a few seconds later, an admin issued a `echo 3 > /proc/sys/vm/drop_cached` as their count dropped.
|
