Introduction
NOCtool will be a distributed system and network monitor. At the
moment, the implementation has graphing support (currently using the
Image library),
persistent storage of collected data, the start of a configuration
language and a few monitor classes.
Supported monitoring
- Disk monitoring for linux hosts
- Ping RTT monitoring
- Load monitoring for linux hosts
- TCP-based application probing (essentially "connect to TCP
socket, send data, read response).
- HTTP monitoring (built on top of the TCP application monitor)
Design
NOCtool monitors are built around a "score" concept, where any
specific monitor has an alert level starting at 0, with higher
values indicating a more severe condition.
A monitored network element is associated with one or more
monitors and by default the alert level for the piece of equipment
is the maximum of any of the monitors attached to it.
NOCtool is designed to have multiple instances running, with data
being fed back from front-line monitoring hosts to back-end
displays.
Current status
NOCtool is currently a work in progress and not really suitable for
deployment. There is web access to the CVS repository and anonymous cvs via cvs -z3 -d :pserver:anonymous:anonymous@common-lisp.net:/project/noctool/cvsroot co source.
There's some guidance to NOCtool internals, for those who are interested.