Hello everyone, I have been looking into the most preferred(with reasons) Application and network monitoring (open source) tools as a combined solution for multiple servers (physical and virtual platforms).
Our requirements include: - Open Source solution (enterprise edition may work at later stage) - Monitors Physical and even virtual platforms - Highly scalable for hundreds of servers - Native Database server: PostgreSQL (+ mysql and oracle) - Role based Alerting - Every bit of monitoring (with I/O cost incurred in a process, availability of resources, Rescue call, automatic redemption etc ) - Easy nail down of an issue (needle in a haystack) - Graphs with dashboard (for all servers running) - Reporting (on daily/weekly basis) - ability to inject our own scripts or monitor custom scripts Some of them are listed below: 1. Nagios <http://www.nagios.org> 2. Zabbix <http://www.zabbix.com/> 3. Hyperic HQ <http://www.springsource.com/products/systems-management> 4. icinga <http://www.icinga.org/> 5. pandoraFMS <http://www.pandorafms.org/> etc Other networking tools that are used in combination of Nagios or zabbix are: 1. monit 2. cacti 3. munin 4. ganglia 5. collectD etc Most of them have issues at scaling as they use RRD files to interact with the server I would love to hear/discuss on what is preferred, what troubles while scaling, which architecture is preferred (Hyperic use agent architecture<http://www.springsource.com/products/systems-management/scalability/architecture>), what will suit to my need ? regards, Akshay gupta _______________________________________________ Ilugd mailing list Ilugd@lists.linux-delhi.org http://frodo.hserus.net/mailman/listinfo/ilugd