NetXMS Support Forum

Has anyone else found these no longer work?  Sorry if I missed a previous post, I searched but found nothing before posting this.

Thanks!

I figured it out from running the netxms server from the command line in debug mode.  I caught that it was running an old database schema... oddly enough, I was able to use the GUI client, just not the web client.  Not sure I have ever had that happen before... either way, the database update resolved my issue.

Thanks for the help!

Thanks for the tip, but still no go.  I also installed on a fresh server and it doesn't work there either, same error (shown below). 

Server uses incompatible version of communication protocol (server= [60, 4, 1, 1, 1,43, 1, 2] client= [60, 4, 1, 1, 1, 44, 1, 2])

This happens with the new UI and the legacy versions.

Anyone? Bueller....Bueller

Thanks in advance

So I have NetXMS server 4.3.2 running, local client connects, no issue.  The WebUI however is not.  It reports: Server uses incompatible version of communication protocol.

The odd part, I am running 4.3.2 on both the server, local client and WebUI, yet it reports version mismatch.  I even downgraded the WebUI to v4.3.1 and I get the same error, but when I looked more closely at the error, the version doesn't seem to match up?  The error is: Server uses incompatible version of communication protocol.  (server= [60, 4, 1, 1, 1, 43, 1, 2] client= [60, 4, 1, 1, 1, 44, 1, 2]). Neither of these seem to match up with the real version... I am expecting to see 4.3.2 or 4.3.1 for the client, but seeing what I can only suspect is 4.3.1.2 (server) and 4.4.1.2 (client).

Any ideas?

Is there any way to add additional passive rack elements, like a fiber patch panel for example.  Would be great to be able to add these to make the system image resemble the rack as much as possible.

I apologize if this has been answered, I couldn't find it in my searches... but I am having issues with network maps.  When setting the background to the exact same zoom level as is displayed using a GEOLOCATION MAP (by right clicking a container for example) my network map does not look anything like the GEOLOCATION map.  The center of the map is way off the canvas. In fact, my center point cannot even be seen.  Secondly, the map does not fill the entire screen.

Is there a way I can make the network map look as good as the GEOLOCATION map? My desire is to have a map that supports network connection linkage, which the GEO map doesn't support.  I will also note, the world map generates fine as well... just the "Network Maps" section is where I have issues.

I have attached the maps below and you can try it for yourself.  Set the geo container to N40 by W85 with a zoom of 10 and then do the same thing with a network map.
You will see in my comparison, the network zooms far south east versus the geo map which centers correctly.

Am I doing something wrong or just better off using an image based map?

Sorry for my delay... I thought I responded.  Thank you! Worked perfectly!

Hello,

I am new to writing scripts for discovery and have that down ok... but I am looking for a way to also filter the interfaces if they are not active.

In this case, I have a sonicwall I wish to monitor, but I don't need to see all the ports not in use.  They all show UP because they use PortShields which connect all the ports in an UP fashion, so I wanted to filter on MTU which in SNMP shows zero on all the ports I wish to ignore.

My SNMP DCI looks like this:

Param: .1.3.6.1.2.1.31.1.1.1.10.{instance}
Discovery (OID): .1.3.6.1.2.1.2.2.1.1

Discovery Filter:
transport = CreateSNMPTransport($node);
if (transport == null)
return false;

// get the interface name to return as instance-name
vb = SNMPGet(transport, ".1.3.6.1.2.1.31.1.1.1.1." . $1);
if (vb == null)
return false;

// filter out loopback
if (vb->value == "lo")
return false;

return %(true, $1, vb->value);

I know MTU is pulled in and stored with the interfaces, but not sure how to filter in the discovery script I have so far.

Like I said, pretty new at this and just getting my hands dirty.... so any help would be greatly appreciated!

I changed to INT and it works fine now... Thanks, I also found a thread on the SNMP walk issue.  Fixed the permission and all good now.

Hi Victor,

These are Unsigned INT64

Also, I am finding (perhaps related) when I do an 'walk' on any SNMP object, I get: "Cannot do SNMP MIB tree walk: Access denied" and it worked before...

A little more info, I tracked down which item it was... it is actually a Sonicwall node, the transformation is a simple ($1 / 1000000) script.... I rebuilt the MIBS and rebooted... doubt that will help.

I am getting an odd transformation error (that never occurred before, until after the 2.2.1 update).  This is happening on the NetXMS server agent node.

(HOSTNAME xxx.xxx.xxx::227::TransformationScript) execution error: Error 13 in line 1: Cannot do automatic type cast

Unfortunately from this error I am not able to determine which of the several data collection items I have is the culprit.  I am getting results back on all my collection items, no invalid data but still getting the error... thoughts?

Thanks in advance!

I looked around and could not find a post about this... from time to time, I get new network interfaces showing up on machines that are not changing in any way.  I have ruled out multiple ip's, nic's etc.  These literally happen all the time and I think to date all in Win OS environments.  Today for example, I have 3 new Local Area Connections (labeled 9,10 and 11).  The server in question has but a single IP and ipv6 is off.... yet each of the NICS being reported as new show ipv6 addresses.  It shows the interface type as tunnel(x3) and one called Interface Type=6 (ethernetCsmacd) which has an IP which isn't even in this subnet -and- belongs to a different server.  Further, this server does not have any VPN services or any other type of tunnel protocol enabled. 

ipconfig/all shows none of these adapters, mac addresses or anything.  Any thoughts as to why this keeps happening and how to stop it?

Thanks!

Victor,

Thanks for the reply.  I did make that change after noticing that myself.  I changed to the SYS_TABLE... however, I still get false positives.  Unless I need to be looking for hex codes versus verbose text comparisons.  This also happens with a table looking at volume sizes... seems like it could be a bug?