Unix

The Technical Specifications below provide details on metrics collected by the Unix application, how those metrics are evaluated by Longitude, and the available preconfigured reports. For additional details on the Longitude Unix application, see the Unix Application User Guide or contact Heroix Support.

Please note that the specific collection commands for metrics will vary based on the operating system being monitored.

Events

Longitude Events are generated when collected performance and availability metrics are evaluated by Longitude’s rule engine and a potential problem is detected.

The rules use best-practice thresholds, displayed below, which can be customized as needed.

Reports

Longitude reports display performance and availability metrics over time. Summary reports display average metrics across multiple devices, while detail reports drill down and present a more granular view of metrics for a selected device.

 

CPU
Metrics Events
CPU user utilization
CPU system utilization
CPU Idle %
Non-Zombie Processes
Runnable processes
Zombie Processes
Cpu iowait%

 

CPU Saturated: % Processor Time > 80%
CPU High: % Processor Time > 60%
Maximum Process Count: Non-zombie Processes > 150
Maximum Runnable Processes: Runnable Processes > 5
Maximum Zombie Processes: Zombie Processes > 0
Reports
Summary CPU usage across multiple servers
Detail Detailed per server CPU usage
Summary Process count across multiple servers
Detail Detailed per server process count

 

File System
Metrics Events
File System
Mount Point
Used Space
Free Space
Used Inodes
Free Inodes
File System Space Critical: Free space % < 2% or
Free space MB < 500 MB
File System Space Low: Free space % < 4% or
Free space MB < 1000 MB
Minimum Free File System Inodes: Free inodes % < 20% or
Free Inodes < 1000

 

Reports
Summary File System Storage across multiple servers
Detail File System Storage for all disks on a server

 

IO
Metrics Events
Device
Disk IO rate in KB

 

Maximum Disk IO Rate: Disk IO Rate < 4000

 

Reports
Summary Disk IO Rate across multiple servers
Detail Disk IO Rate for all disks on a server

 

Memory
Metrics Events
Page outs/second
Pages swapped out
Total memory in kilobytes
Free memory in kilobytes
Total swap space in kilobytes
Used swap space in kilobytes
Pages scanned/second

 

Minimum Free Memory: Free Memory < 5%
Maximum Page Out Rate: Page outs/second > 5000
Maximum Page Scan Rate: Pages scanned/second > 500
Maximum Swap Out Rate: Pages swapped out > 1
Critical Swap Used %: Used swap space > 90%
High Swap Used %: Used swap space > 80%

 

Reports
Summary File System Storage across multiple servers
Detail File System Storage for all disks on a server

 

Network
Metrics Events
Interface Name
Packets Received
Packets Transmitted
Bad Packets Received
Bad Packets Transmitted
Collisions

 

Maximum Packet Collision %: 100 * Collisions/(Packets Received + Packets Transmitted) < 5
Maximum Interface Input Error Rate: sum(Bad Packets Received)/(Collection Period) <= 0
Maximum Interface Input Rate: sum(Packets Received)/(Collection Period) <= 6250
Maximum Interface Output Error Rate: sum(Bad Packets Sent)/(Collection Period) <= 0
Maximum Interface Output Rate: sum(Packets Sent)/(Collection Period) <= 6250
Reports
Summary Packets sent, received and total across multiple servers and interfaces
Detail Detailed Packets sent, received and total per server and interface

 

Network Interface
Metrics Events
Name
Output Queue Length
Packets Output Errors
Packets Received Errors
Kilobytes Received/Sec
Kilobytes Sent/sec

 

Input Error Rate: Packets Received Errors > 0
Input KB Rate: KB Received/sec > 12,500
Output Error Rate: Packets Output Errors > 0
Output KB Rate: KB Sent/sec > 12,500
Reports
Summary Input rate across multiple servers
Detail Detailed per server input rate
Summary Output rate across multiple servers
Detail Detailed per server output rate
Summary Input + output rate across multiple servers
Detail Detailed per server input + output rate

 

Processes
Metrics Events
Command with Arguments
Command
Process ID
Parent Process ID
User Name
CPU Time (ms)
Elapsed Time (s)
Terminal
Memory Pct

 

CPU Hogging Process: (Process CPU Time)/(Collection Interval * #Processors) > 50%
High CPU Process: (Process CPU Time)/(Collection Interval * #Processors) > 30%
High Memory Process: Process Memory Pct > 20%

Top
Transactions VMware