NetXMS Support Forum

I've had NetXMS core crash on me once so far and it won't restart because the database is locked, this requires running "C:\NetXMS\bin\nxdbmgr.exe check" but that requires manually answering yes and then later confirming to save changes

Is there a parameter to automate this, so it automatically force unlocks the database and commits changes. That way I can write a simple .bat script to run and check if netXMS is not running and do the above
Because we don't use netXMS constantly we can't really afford for it to crash and nobody is aware for several hours (or days if its a weekend) and we lose a huge amount of logging during that period. Hence the need for an automated script to get it up and running again as quickly as possible

We're a managed services company and have decided to use NetXMS as our network monitoring software. I'm still in the early stages of deployment and getting to understand how NetXMS works
Something i'm a bit confused on is in regards to network discovery and agents
We typically don't have a VPN into a clients network and so we can't ever get direct access to their subnet from our network (and we have a lot of clients that use the same subnet range)

We want to install an Agent in their network, and have that agent do all the polling and network discovery of their internal network. So that we don't have to manually add every device 1 by 1

What I don't understand is firstly..

1) How does NetXMS perform its network discovery? In Configuration -> Network Discovery I enter IP ranges. Are these ranges ONLY scanned by the server that NetXMS server is install on? Or does the server tell all the agents "Hey can you please scan for these configured IP ranges and report back to me"
If it's the former, then how do I get an agent to do active discovery?
If its the latter, is it possible to restrict the scope for each agent. Since I know that most clients only have a single class C network so it would be a huge waste of time scanning 172.16.0.0/12. I would want to tell each agent to only scan on its relevant subnets

2) I believe at this point in time, Agents are polled by the server and not the other way around, correct? So we have to port-forward on the client/Agent's side
Ideally I would like to be able to install an agent in their network, tell it our IP address and it will send to us. So that we don't have to port-forward on every clients network. Is this possible?

3) How do I add devices and have the remote agent be the one that does all the polling. I'm trying to poll a HP switch with a remote agent and its just not working. What exactly do I have to do
What I've done is 'Create Node' -> Set IP address to that of the HP Switch, set 'Proxy for SNMP' to the remote agent and clicked ok. Is this it? Do I need to do anything else? Cause its not working. This is what I get
[14.10.2016 15:01:32] **** Poll request sent to server ****
[14.10.2016 15:01:32] Poll request accepted
[14.10.2016 15:01:32] Starting configuration poll for node SW-Core
[14.10.2016 15:01:32] Capability reset
[14.10.2016 15:01:32] Checking node's capabilities...
[14.10.2016 15:01:32]    Checking SNMP...
[14.10.2016 15:01:33] Capability check finished
[14.10.2016 15:01:33] Checking interface configuration...
[14.10.2016 15:01:33] Unable to get interface list from node
[14.10.2016 15:01:33]    Interface "unknown" is no longer exist
[14.10.2016 15:01:33] Interface configuration check finished
[14.10.2016 15:01:33] Checking node name
[14.10.2016 15:01:33] Node name is OK
[14.10.2016 15:01:33] Finished configuration poll for node SW-Core
[14.10.2016 15:01:33] Node configuration was not changed after poll
[14.10.2016 15:01:33] **** Poll completed successfully ****

Remote agent has port 4700 forwarded, but not 161. Does the master server send all commands to the agent on port 4700 or do I also have to open up 161 (or other ports?)

I've seen this mentioned in the manual under Instance Discovery
NetXMS can create parameters for Data Collection Item automatically

But with no mention of how to do it. So.... how do I have it so all interfaces are automatically added for monitoring. And perhaps additional intelligence so it knows if it's a MikroTik to add the MikroTik templates etc
Even better, can it try multiple templates and then automatically remove all the ones that are unsupported?