21 lines
1.2 KiB
ReStructuredText
21 lines
1.2 KiB
ReStructuredText
`Home <index.html>`_ OpenStack-Salt Installation Manual
|
||
|
||
Chapter 1. Overview
|
||
-------------------
|
||
|
||
.. toctree::
|
||
|
||
The overall health of the systems is measured continuously. The metering system collects metrics from the systems and store them in time-series database for furher evaluation and analysys. The log collecting system collects logs from all systems, transforms them to unified form and stores them for analysis. The monitoring system checks for functionality of separate systems and raises events in case of threshold breach. The monitoring systems may query log and time-series databases for accident patterns and raise an event if anomaly is detected.
|
||
|
||
.. figure :: /operate/figures/monitoring_system.svg
|
||
:width: 100%
|
||
:align: center
|
||
|
||
**The difference between monitoring and metering**
|
||
|
||
Monitoring is generally used to check for functionality on the overall system and to figure out if the hardware for the overall installation and usage needs to be scaled up. With monitoring, we also do not care that much if we have lost some samples in between. Metering is required for information gathering on usage as a base for resource utilisation. Many monitoring checks are simple meter checks with threshold definitions.
|
||
|
||
--------------
|
||
|
||
.. include:: navigation.txt
|