Node Data Collection

Contributors netapp-alavoie

Cloud Insights gathers metrics from the node on which you install an agent.

Installation

  1. From Admin > Data Collectors, choose an operating system/platform. Note that installing any integration data collector (Kubernetes, Docker, Apache, etc.) will also configure node data collection.

  2. Follow the instructions to configure the agent. The instructions vary depending on the type of Operating System or Platform you are using to collect data.

Objects and Counters

The following objects and their counters are collected as Node metrics:

Object: Identifiers: Attributes: Datapoints:

Node Filesystem

Node UUID
Device
Path
Type

Node IP
Node Name
Node OS
Mode

Free
Inodes Free
Inodes Total
Inodes Used
Total
Used Total
Used

Node Disk

Node UUID
Disk

Node IP
Node Name
Node OS

IO Time Total
IOPS In Progress
Read Bytes (per sec)
Read Time Total
Reads (per sec)
Weighted IO Time Total
Write Bytes (per sec)
Write Time Total
Writes (per sec)
Current Disk Queue Length
Write Time
Read Time
IO Time

Node CPU

Node UUID
CPU

Node IP
Node Name
Node OS

System CPU Usage
User CPU Usage
Idle CPU Usage
Processor CPU Usage
Interrupt CPU Usage
DPC CPU Usage

Node

Node UUID

Node IP
Node Name
Node OS

Kernel Boot Time
Kernel Context Switches (per sec)
Kernel Entropy Available
Kernel Interrupts (per sec)
Kernel Processes Forked (per sec)
Memory Active
Memory Available Total
Memory Available
Memory Buffered
Memory Cached
Memory Commit Limit
Memory Committed As
Memory Dirty
Memory Free
Memory High Free
Memory High Total
Memory Huge Page Size
Memory Huge Pages Free
Memory Huge Pages Total
Memory Low Free
Memory Low Total
Memory Mapped
Memory Page Tables
Memory Shared
Memory Slab
Memory Swap Cached
Memory Swap Free
Memory Swap Total
Memory Total
Memory Used Total
Memory Used
Memory Vmalloc Chunk
Memory Vmalloc Total
Memory Vmalloc Used
Memory Wired
Memory Writeback Total
Memory Writeback Tmp
Memory Cache Faults
Memory Demand Zero Faults
Memory Page Faults
Memory Pages
Memory Nonpaged
Memory Paged
Memory Cache Core
Memory Standby Cache Normal
Memory Standby Cache Reserve
Memory Transition Faults
Processes Blocked
Processes Dead
Processes Idle
Processes Paging
Processes Running
Processes Sleeping
Processes Stopped
Processes Total
Processes Total Threads
Processes Unknown
Processes Zombies
Processor Queue Length
Swap Free
Swap Total
Swap Used Total
Swap Used
Swap In
Swap Out
System Uptime
System Num CPU
System Num Users
System Calls

Node Network

Network Interface
Node UUID

Node Name
Node IP
Node OS

Bytes Received
Bytes Sent
Packets Outboud Discarded
Packets Outboud Errors
Packets Received Discarded
Packets Received Errors
Packets Received
Packets Sent

Setup

Setup and Troubleshooting information can be found on the Configuring an Agent page.

MacOS Memory Usage

Cloud Insights (via Telegraf) and macOS report different numbers for memory usage. Both Telegraf and the Mac activity monitor use metrics gathered from vm_stat, however the total memory usage is calculated differently for each.

Telegraf calculates Memory Used Total as follows:

Memory Used Total = Memory Total - Memory Available Total

Where Memory Available Total is derived from the sum of "Pages free" and "Pages inactive" in vm_stat.

The Mac activity monitor, on the other hand, calculates Memory Used as follows:

Memory Used = App Memory + Wired Memory + Compressed

Where:

  • App Memory is derived from the difference between “Anonymous pages” and “Pages purgeable” in vm_stat,

  • Wired Memory is derived from “Pages wired down” in vm_stat, and

  • Compressed is derived from “Pages occupied by compressor” in vm_stat.