Hello.
As one of the NetXMS developers I want to clarify situation a bit.
First of all - NetXMS is quite mature now; our first goal is achieved - system is stable, capable of monitoring large networks (hundreds of nodes and thousands of data collection items), highly configurable and and easy to maintain (e.g. agent configuration can be edited from administrator's management console; same for upgrades - you don't have to manually upgrade every box, it can be done from management console).
As for features - we support data collection using SNMP v1/v2 (including traps) and our own agent (ported to most common platforms, as well as Netware and IPSO). Our own trap mechanism is also available for pushing data and events directly into the server. All NXCP-based communications are encrypted and authenticated. Agent can proxy NXCP/SNMP requests, so networks behind NAT can be easily monitored.
Agent provide range of parameters and can be easily extended for specific needs (e.g. we use custom subagent to monitor ATMs). Core agent on all platforms provide cpu/network/process/file information; OS-specific/task-specific features provided by subagents (windows - PDH and WMI, Solaris - kstat, etc, subagent for UPS monitoring, etc). All agents can also use external applications/scripts as data source.
Our biggest problem right now is outdated documentation; it's prioritized but not done yet. However, is you have any question - feel free to ask here or in our forum (NetXMS Support Forum - Index)
wbr, alex.
