NetXMS Support Forum

From time to time, I find myself trying to locate "objects that already exists". This is becoming time consuming. When adding the host, the error message only displays that the object already exists but shows no additional information. So I am lost as to exactly why the error is displayed.

This happened just now when I tried to register a "foobar.domain.com" and I get the error message. I know for a fact that "foobar" was never registered in netxms before (new naming standard) but perhaps the IP was used on an old object that was removed. I tried registering "foobar" with both dns and ip - same result every time.

It would be a major improvement if the error message could show some additional details about the duplicate object.

Hi

This is not really general support, so please forgive me as I didn't find a better place to put this.

I find it puzzling that the beta is being pushed as the latest download, and latest stable (1.2.17 ?) is found in the archive.

The number one priority for me as a user of netxms is a reliable and robust alerting system. Ofcourse I want to contribute on testing the latest code, but not in production. So I am a bit confused as to why the 2.0 branch is promoted on the download page, and 1.2.17 is "hidden".

I would like to see the download page split into "stable" and "milestone/testing". It would be a step in the right direction.

I'm having a problem with the android client. I've setup a few custom graphs, which have colors set to "auto". This works very well with the desktop nxmc client, however with the android client all lines are white against a black background.

This makes the data impossible to differ, and I have to make a best guess on which is what.

A work-around is to manually set the colors on the desktop client which is then reflected in the android client - but that quickly becomes a hazzle with multiple graphs.

Discovered something weird today. I configured a threshold alarm, which should trigger an alarm if a given string is not correct after 3 attempts.

See the picture attachment. The condition is triggered, but no alarm on the host is activated.

Is there a logical reason why it fails? I have other similar thresholds, but they are integer and float. So my theory is that it has something to do with the check being a string.

I've noticed in the last couple of releases that the check I have for disk space percentage shows up as "0". However, If I generate graphs the real values are plotted correctly. Is it just me or is it occurring for others as well?

After upgrading to the latest version of netxms, I've experienced two cases of the netxms-server process dying.

The only error is found in /var/log/messages;

Apr  4 09:52:07 netxms kernel: [6647577.566437] netxmsd[29596]: segfault at 0 ip 00007f2a203a8f96 sp 00007f2a152bc9e0 error 4 in libnetxms.so.1.0.0[7f2a20382000+39000]

Please advice. I'll throw in a cron ps-check in the meanwhile.

We have a few hosts that are not capable of running netxms agents. VMware ESX for instance.

Is there a simple way of utilizing a netxms agent on a remote host that does cli checks on the ESX hosts, and tie these values to the ESX host in netxms?

Currently I have the ESX values stored on the host that is running the netxms-agent, along with its own checks. Not a pretty sight. I tried the proxy functionality but I guess that's just for relaying agent traffic.

Seems like <<ERROR>> values are not raised by netxms. Not out of the box anyway.

I'd like to get some sort of warning when checks fail. Any tips would be greatly appreciated.

Hi

I know this has been discussed before without any progress, so I'm raising the issue again.

Today's Object Tools has tons of potential, but in a diverse environment it generates noise as every entry is shown on _all_ monitored hosts.

Would it be possible to have some sort of filter or selecting the hosts that qualify for a new entry?

Hi

I'm using netxms to monitor our fw. However when someone uses vpn, a new ppp adapter appears. This causes a notification in netxms which is fine. However when I try to <IGNORE> the adapter, it respawns with the next vpn user and another notification is shown. It seems that PPP1 one day is not the same PPP1 the next day.

Any tips on how to ignore these adapters are welcome. I'd like to keep the functionality of a notification when a new adapter appears, but ignore the ppp adapters on the fw.

Hi

I need to monitor raids on two raid-controllers, and get notifications if a hard-drive fails.

How can I use netxms to accomplish this?

If it matters, the servers in question are VMware esxi hosts, and the controllers are LSI SAS controllers. The servers also have ipmi, which might present options.

Hi

I'd rather not put the netxms server in DMZ for security reasons, but I would love to use the mobile client outside our private network.

Any input on how to achieve this would be greatly appreciated.

Hi

I can't seem to find a logical way to remove a node from a template.

A quick howto would be greatly appreciated.

Even-though the packages work on Wheezy when porting an older version of libssl from squeeze, I would appreciate if we could get native wheezy packages.

There is also a problem with dependencies, as postgresql-8.4 is not standard in newer debian releases. So each time I install netxms, I manually have to modify the packages and change the dependency from postgresql-8.4 to postgresql-9.1

Squeeze is nearing EOL, so perhaps its time to move on? 

Hi

I'm having mixed feelings when using netxms as I have several times thought that everything was working , and when I view the data from one node it can be marked as red for whatever reason (one example is that the poll is not supported on that node, or the netxms agent is not functioning properly). I suspect that it might be a user error on my part, but has anyone else experienced this and know how to avoid it? I really need to know if a poll has failed. It should trigger a notification.

Hi

One of the biggest issues that holds netxms back is the lack of AD/LDAP feature for authentication.

Are there any plans to introduce this functionality soon?

Hi

I have been trying to configure remote actions on agent host.

What I am trying to accomplish is to restart a service if it stops responding. The service is called Varnish.

I have defined an action like this in the nxagentd.conf

Action = VarnishReload: service varnish restart

And trying to execute it by calling this on the server:

Action:VarnishReload

The netxms server log shows the following

[11-Jun-2012 20:47:35] *actions* Executing on "78.XX.XXX.XXX": "Action:VarnishReload"

However, nothing happens. What am I doing wrong?

Hi

I`ve had a look at installing the nxagent on my dd-wrt flashed router which runs linux. Since there are noe ipkg packages available, im a bit at a loss. Has anyone done anything like this before, or can point me to a ipkg package?

Hi

After installing netxms-agent_1.2.0_amd64.deb all seems well until reboot where the agent does not start as a normal service.

I have checked for /etc/default/ settings in case the agent needs to be told to start on boot, but I have not found anything.

The systems in question are debian squeeze, and ubuntu 10.04 LTS.

Any ideas on what`s wrong?