Monitor

Getting started

New Install

Reference

Guides

Monitor

Every three seconds creative.space collects over 2,000 metrics, that can be presented as hundreds of charts, and is currently monitored by approximately 150 potential alarms

System Highlights

A digest of key information pulled from across the system metrics.

Charts (From Left to Right):

RAM - Percentage of system memory currently being utilized.
Network In - The current bitrate of data being pushed to the server.
Network Out - The current bitrate of data being pulled from the server.
CPU - Percentage of total CPU capacity currently being utilized.
Disk Write - Bytes currently being written to disk.
Disk Read - Bytes currently being read from disk.
Uptime - How long the hardware has been running since last power off.

CPU

Metrics related to the Central Processing Unit (CPU) for your node.

Charts (In Order):

Usage - Total CPU utilization (all cores). 100% here means there is no CPU idle time at all. If ‘iowait’ is constantly high, your disks are a bottleneck and they slow your system down. A constantly high percentage of softirq may indicate network driver issues.
Load - Current system load, i.e. the number of processes using CPU or waiting for system resources (usually CPU and disk). The 3 metrics refer to 1, 5 and 15 minute averages. The system calculates this once every 5 seconds. Numbers below 1.00 represent the system load is not at 100% i.e. a number of .6 shows that the load is 60% and an idle time of 40% on average. A number of 1.73 shows the system was overloaded by 73% on average.
Temperature - The temperature of the CPU(s) Cores. Maximum operating temperature is 194° F (Fahrenheit) or 90° C (Celsius).

Memory

Metrics for the physical memory in your node, which is utilized heavily by ZFS.

Charts (In Order):

RAM Usage - System Random Access Memory (i.e. physical memory) usage.
RAM Available - Amount of RAM available for usage.

Network

Metrics related to your networking interfaces.

Charts (In Order):

Aggregated Bandwidth - Total bandwidth of all physical network interfaces. This does not include lo, VPNs, network bridges, IF devices, bond interfaces, etc. Only the bandwidth of physical network interfaces is aggregated.
IP Bandwidth - Total IP traffic in the system.
IPv4 TCP Sockets - Total IPv4 sockets being used.
IPv4 UDP Packets - Total IPv4 UDP packets.
IPv4 Packets - Total IPv4 packets per second received, send, and delivered on the system.
IPv4 Errors - Total IPv4 errors on the system.
IPv6 Packets - Total IPv6 packets per second received, send, and delivered on the system.
IPv6 Bandwidth - Total IPv6 traffic in the system.

Disk Usage

Metrics related to I/O for your physical drives.

Users

Metrics related to users on the system and related activity.

Charts (In Order):

CPU Time - Total CPU time per user, including system users for the database and other processes. Each CPU core is measures as 100%, i.e. 8 CPU cores = 800%.
Open Files - Total amount of files open per user, including system processes.
Disk Reads - Current actual disk reads in bytes per user.
Disk Writes - Current actual disk writes in bytes per user.
Disk Logical Reads - Current reads in bytes per user, including reads from the cache.
Disk Logical Writes - Current writes in bytes per user including initial writes to cache.

ZFS

The ZFS filesystem utilizes an Adaptive Replacement Cache (ARC) to enhance performance. This cache leverages server memory (RAM) to deliver superior performance compared to traditional spinning disk RAID systems.

By predicting the data that will be accessed, the need for actual disk reads is minimized, as the data is cached.

When data is written, it is first stored in memory and then written to the disk, with each block undergoing checksum verification.

The charts for ZFS provide real-time insights into the ARC and it’s efficacy.

Charts (In Order):

ARC Size - The amount of RAM currently used for the ARC is displayed.
Reads - The number of read operations per second.

‘ARC’ is the RAM cache.
’Demand’ is the opposite of prefetch: direct requests to the ARC, not predicted requests.
‘Prefetch’ is the ZFS read-ahead feature, to predict and pre-cache blocks for streaming (sequential) workloads.
‘Metadata’ describes the ZFS dataset (file system or volume) and the objects within it. The data is the contents of those objects, including file, directory and volume blocks.

ARC Size Breakdown - The percentage of recent and frequent files stored in cache. This percentage may dynamically change based off user activity.
ARC Hits - The ARC "hit" ratio/rate represents how frequently the requested data is found in the cache. When it is not found, it results in a cache "miss", requiring the data to be retrieved from disk (which is slower). Monitoring the number of hits can provide valuable insights into the efficiency of your caching.
Demand Hits - The percentage of demands that are being hit or missed.

Processes

All the current processes running on the CPU and what percentage is running or blocked.

← Previous

System

Interfaces

On this page

Monitor
System Highlights
CPU
Memory
Network
Disk Usage
Users
ZFS
Processes