Network Active Discovery and multiple Subnets

[victor]

Hi,

seems to be a bug in the server. What NetXMS  version you are using?

Best regards,
Victor

[jreach]

Hello,

Thank you for the response.

I am using version 2.0.8 and it is installed on a Windows 2012R2 x64 Server. The database is on a separate server (Server 2008R2 x64 / MSSQL 2008 R2). An ODBC Connection is used to connect to the Server/SQL Database.

[jreach]

Anyone have any suggestions on what I can try, to get this working correctly?

[victor]

Hi,

try to upgrade to 2.1 if possible.

Best regards,
Victor

[jreach]

Hi,

I will do that.

When I go to upgrade. What are the exact steps to follow (on Windows Server)?

To stop the NetXMS server, do I just stop the services or is there a special command line command I need to run to stop it?

[lindeamon]

from the netxms management console:
tools->server console

then, invoke the down command. this will close the management console and start shutting down the server.
then, stop both netxms core and netxms agent services.
finally, run the installer and it will upgrade what you need.

troubleshooting:
if the server will not start after the upgrade, there is a chance that the DB wasn't upgraded so:

1. open command line
2. go to <netxms's installtion>\bin folder. default is: c:\netxms\bin
3. run nxdbmgr.exe upgrade
4. cross your fingers 

Good Luck

[jreach]

Hi All,

I updated to v2.1 but I am still seeing the same issue as originally posted. Do you have any suggestions?

[jreach]

Well I completely deleted the database and uninstalled NetXMS.

I then re-installed a fresh copy and re-configured all the Network Discovery Settings.

How do I restart the NetXMS server correctly? It is running on a Windows Server. Do I just reboot the Windows Server or is there a command I can run from the Windows Server command line to restart the NetXMS server?

[jreach]

Alright,

So it does look like the clean re-install has fixed the multiple subnet issue, to a degree.

Now, I am seeing multiple subnets showing up in the 'Entire Network' tree.

I also have 'E-mail Alerts' set up for any node down/up events. And, I am getting Down/Up events/e-mails, for some subnets that are not showing up in the tree on the left at all (but they are in the database). Any idea on what could be causing that?

Also, in the database, these missing IPs have a netmask of 32. When it should be 24 (also my discovery auto filter is set for each subnet as such AAA.BBB.CCC.0/24 or AAA.BBB.0.0/16).

[jreach]

Alright, so I jumped the gun on the discovery working correctly. ActiveDiscoveryInterval is set at 7200 (default) and it has been approx. 24 hours since I did a clean re-install of everything.

We have multiple subnets.

Subnet A: Found Everything as expected
Subnet B: Found Everything as expected
Subnet Found Everything as expected
Subnet X+: Nothing being discovered

Example:

Total of 40 subnets in the Active Discovery Section on the Network Discovery Configuration Page.
For the Address Filters I have Automatically Generated script with following rules: Accept node if it is within given range or subnet.

I am not using any subnets in the filter section, I am only using address ranges. I am doing this because for some reason nodes were being discovered with a 32 netmask when they should be 24, and this would cause them not to be placed in the Tree on the left (under Entire Network).

*Below subnets are not actual subnets, but correctly depict format.

Address Filter Range Example:

• 10.10.10.1 - 10.10.10.99
• 10.10.10.150 - 10.10.10.254
• 10.10.11.1 - 10.10.11.254
• 10.10.XX.1 - 10.10.XX.254 (XX ranges from 12 to 30)

Active Discovery Target List:

• 10.10.10.0/24 - Discovered
• 10.10.11.0/24 - Discovered
• 10.10.12.0/24 - Discovered
• 10.10.13.0/24 - Discovered
• 10.10.14.0/24 - Discovered
• 10.10.15.0/24 - Discovered
• 10.10.16.0/24 - Discovered
• 10.10.17.0/24 - Discovered
• 10.10.18.0/24 - Discovered
• 10.10.19.0/24 - Discovered
• 10.10.20.0/24 - Discovered
• 10.10.21.0/24 - Discovered
• 10.10.22.0/24 - Discovered
• 10.10.23.0/24 - Discovered
• 10.10.24.0/24 - Discovered
• 10.10.25.0/24 - Discovered
• 10.10.26.0/24 - Discovered
• 10.10.27.0/24 - Discovered
• 10.10.28.0/24 - Discovered
• 10.10.29.0/24 - Discovered
• 10.10.30.0/24 - Discovered
• 10.11.0.0/16
  • 10.11.10.0/24 - Discovered
  • 10.11.20.0/24 - Discovered
• 10.XX.0.0/16 (XX goes from 12 to 30)
  • 10.XX.10.0/24 - Not Discovered
  • 10.XX.20.0/24 - Not Discovered

[jreach]

Alright, so I got the server to start finding all the subnets. I had to change my active discovery subnets from the  /16 subnet, to two /24 subnets. I guess a /16 is just too large an address space for the server to search through in a timely fashion (especially since most of the subnet space it was looking at, was non-existent).

The only issue that is left is that not a single node that it discovers, does it discover it with the proper subnet. It is finding everything as a /32 instead of /24...so right now what I have to do is go in and stop the server, manually change the subnets in the database, then start the server back up. At this point all the hidden nodes start showing up and going in their correct groups.