NetXMS Support Forum

Hi,

I'm having a small issue in the Win x64 NetXMS v1.2.13 console.

Sometimes, if I add a new custom attribute on a node, close the properties and re-open them, it appears to be gone.
When trying to add the attribute again, it doesn't give an error or anything, but still disappears again after closing the properties window.

After closing and re-opening the console the newly created attributes reappear and everything works normally again.

Maybe the behavior only pops up after the console has been open some time.

Any clues?

Kind regards,
bdefloo

Hi,

I was playing around with the max memory settings for the v1.2.11 NetXMS Console when I noticed something peculiar:

My console is usually using around 240MB out of ~600MB allocated heap space (as seen in the status bar of the console). If I reduce the -Xmx parameter from 1024m to 512m This is 240MB out of ~450MB.

I tried reducing it further to 256m, this caused an error to pop up saying it failed to load the MIB file, but the console then only used 60MB! Further tests reveal that if I rename the mib file on the server, and start the console with an adequate heap size (eg 1024MB), only 80-100MB is used.

Is the MIB file integrally loaded and expanded into the heap every time the console is started? My MIB file is only 13MB on disk, but appears to account for 100+ MB of used heap space.

If so, isn't this a bit wasteful of resources? After all, I can't imagine most people use the MIB browser daily. Maybe it could be loaded when opening the MIB browser tab?

Thanks for your input!

Kind regards,
bdefloo

Hi,

I have NetXMS x64 and MS SQL 2012 Express x64 on a Windows 2008 R2 server.

Hi,

Since upgrading to v1.2.11 I've been getting more "Request timed out" errors in the management console when doing various changes. (applying templates, creating a container, moving nodes, ...)

It doesn't always happen, andsometimes the operation goes through anyway, other times I have to do it again or it only succeeds partially (eg 3 out of 20 nodes are in the template, the others aren't)

Has something changed that could be causing resource locks or delays?

Thanks!

Kind regards,
bdefloo

Hi,

Is there any way to search for a node based on the serial number listed under the "Components" tab in Object Details?

I've tried via the database but can't seem to find the table the components are stored in.

Thanks!

Hello,

I'm still experiencing this issue in v1.2.8 (Windows x64 console).

Have you been able to look at it?

Thanks!

Hi,

You don't need to assign a MIB to a node for it to work.

The NetXMS console shows all the MIBs it knows in it's MIB browser view. Most likely, you'll find what you're looking for somewhere under:
[root]->iso->org->dod->internet->private->enterprises->cisco->ciscoMgmt
You can also enter a OID (.1.3.6.1.4.1.9.9. ...) in the OID field to navigate quicker.

Good luck!

Hi Victor,

I've put the new driver in our NetXMS server, it detects RFS6000 switches nicely and displays the APs. Thanks for the help!

One small additional remark: the AP model is detected as "UNKNOWN" for our AP650 devices.
I suspect this is because of OID
.1.3.6.1.4.1.388.14.3.2.1.11.5.1.5
which for an AP650 has value 9, whilst AP300's are 3.

Kind regards,
bdefloo

Hi,

Yes, I have an image as map background.

I redownloaded the console from the website (nxmc-1.2.7-win32-x64.zip) and ran it on a Windows 7 x64 PC, but the problem still persisted.
Java version:
C:\Windows\system32>java -version
java version "1.6.0_25"
Java(TM) SE Runtime Environment (build 1.6.0_25-b06)
Java HotSpot(TM) 64-Bit Server VM (build 20.0-b11, mixed mode)

I'm afraid I haven't seen any predefined maps so far... It'd find it a really useful features on custom network maps too, though.

Kind regards,
bdefloo

Hi,

I'm having a small problem with maps in the NetXMS v1.2.7 Windows console (x64).

If I open a map and click on a node, everywhere I right-click I get the menu for that node. I can't get back to the menu for the map itsself without closing and reopening the map.
In previous versions if you right-clicked the background you got the map menu.

Might be a bug?

While on the topic of maps; I notice
https://www.radensolutions.com/chiliproject/issues/204#note-3
is marked as resolved. Where can I find this option?

Kind regards,
bdefloo

Hi,

The NetXMS server should do this automatically with the SNMPTrapLogRetentionTime server configuration variable. I don't think there's a way to do this manually at the moment except for throwing the records out in the SQL database.

Kind regards,
bdefloo

Hi,

I'm running NetXMS on Windows Server 2008 R2 (x64).
I've sent SNMP dumps of both tables from a WS5100 and an RFS6000 wireless switch to [email protected].

Like I mentioned, it does happen that an AP is up, but some of its radios are down/unadopted. So being able to view individual radio statuses would be an advantage. (id 62 in the WS5100 dump file is such an example)
Radios are layer 2 interfaces on the access ports, so listing them under their respective AP makes sense. Technically, they also have an ethernet interface with the MAC address of the AP itsself.

One remark is, how will you display (unadopted) radios on an AP that is down (not listed in SNMP)? You could fall back on the AP MAC address which is listed under .1.3.6.1.4.1.388.14.3.2.1.11.5.1.2, and use information from .1.3.6.1.4.1.388.14.3.2.1.9.2 to fill in the name and other details when available.

Kind regards,
bdefloo

Hi,

Since I upgraded to v1.2.7 yesterday I've been having some small problems with creating nodes, containers, templates, template groups, etc.
Some times (not always), the operations takes up to 10-30 seconds, or even gives a message such as "Request timed out". However, if I manually refresh the object browser a while later, the requested operation is usually done.

The NetXMS server is running on Windows 2008 R2 (x64) with MS SQL Express 2012, with 16GB of RAM and two quad core CPU's. CPU usage of netxmsd is about 1-2%, 400MB RAM, sqlservr about 1-3%, 4GB RAM in use.

The queue lengths are all within reason, except for "Database writer's request queue (other queries) for last minute" which sometimes goes up to 1K-2K+. I have 8 database writers configured, and there are 933 nodes with 18296 objects and 8350 DCI's.

Any ideas?
Thanks in advance!

Hi,

Walk results on those OIDs look good! I didn't get any results for unadopted APs, but our wireless switches are set to automatically adopt so there never are any.

The RFS6000 OID (read from .1.3.6.1.2.1.1.2.0) is .1.3.6.1.4.1.388.16. If you want to test before making any final changes to the device driver, feel free to send me a modified version.

There is also a small sidenote I'd like to make: if an AP goes down, it will dissappear from both the adopted and unadopted lists. This makes it difficult to monitor when one goes down, unless you look at the list of access port radios on the wireless switch:
OID .1.3.6.1.4.1.388.14.3.2.1.11.5 lists all configured radios,
.1.3.6.1.4.1.388.14.3.2.1.11.11 lists all adopted radios and their status. So, radios that are down are only listed in the former. In the WS5100/RFS6000 web interface radios that are down are shown with a red cross, which makes them easy to spot. The first list however does not appear to have a status field

Downside to working with radios rather than APs is that a single AP may have multiple radios (eg an 802.11a and a 802.11b/g radio), so they're listed more than once. On the other hand, I have had issues with one of multiple radios being down on a single AP, so this can be a valuable monitoring advantage as well.

Thanks for the support!

Hi,

First of all, great work the new update!
I love the monitoring on our WS5100 wireless switches.

However, we also have a number of motorola RFS6000 switches in use. It'd be great if we could monitor these in the same way. These are the successors to the WS series, and appear to use the same MIBs.

But to be sure, is there any way i can force the SYMBOL-WS driver on such a switch to test? Would you prefer I give you SNMP walk results? (of what OIDs?)

Thanks in advance,
bdefloo