NetXMS Support Forum

Hi!

Expected interface state can be set in interface object's properties, on "Polling" tab. There are 3 possible values:

UP (default) - interface expected to be up. If not, interface object goes to CRITICAL state and SYS_IF_DOWN event generated.
DOWN - interface expected to be down. If it is up, interface object goes to CRITICAL state and SYS_IF_UNEXPECTED_UP event generated.
IGNORE - interface status is of no interest. Interface object will always be in NORMAL state, regardless of link status.

Please note that administratively down interface will still be placed into DISABLED state, regardless of expected state setting.

Best regards,
Victor

Hi!

Windows 2000 is not supported by 1.2.0 agent. I got unexpectedly high number of complains regarding this, so I'll check if it is possible to made agent compatible with Win2000 again. Currently, you can use agent version 1.0.13 for Windows 2000. It's available here: https://www.netxms.org/download/archive/agent_packages/nxagent-1.0.13.exe.

Best regards,
Victor

Hi!

It's definitely should not be like this. I'll try to reproduce this on my test system.

As for bug tracker, we use Jira internally, but because we have very limited number of licenses, we cannot open it to the public. And public bug tracker is very inactive, but I see those issues and will fix them. I have added 1.2.0 and 1.2.1 versions to public bug tracker.

Best regards,
Victor

В WinPerf субагенте есть возможность использования английских названий счетчиков, но только для параметров, определенных в субагенте (через Counter =). Я добавлю в TODO сделать то-же самое для PDH.CounterValue.

Hi!

Currently you can do this using PING subagent. You can load it on agent on management server itself, and configure Icmp.Ping parameters for all nodes you want to ping. If you wish to have those parameters on each target node and not on management server node, you can do this by configuring Icmp.Ping parameters on target nodes and setting proxy node to management server node.

Best regards,
Victor

Это довольно странно - протокол используется один и тот-же и в старой консоли, и в новой. А когда ошибки бывают чаще - при открытии графика или при обновлении?

Hi!

Yes, it's problem in version 1.2.0. It is fixed in upcoming 1.2.1 release.

Btw, why you still using legacy console?

Best regards,
Victor

Quote from: dolph2005 on May 04, 2012, 03:49:33 PM
В дата Data Collection Configuration, не могу создать новый параметр, при выборе NetXMS agent список доступных параметров пуст.

Ne moglu bit' takogo, chto dlja dannogo hosta ne proshel esche configuration poll? Ili eto postojanno i dlja vseh hostov?

Manual budet, mi perenosim suschestvujuschij manual v wiki i potom budem ptravit' ego dlja versii 1.2.0.

Hi!

Better point for such function could be configuration poller. That way you will know that all communication settings are discovered, and getInterfaceList will not fail if node has an agent. Calling getInterfaceList before first configuration poll will always return error. This is why you get errors.

You should not call lock around getInterfaceList calls.

You don't need to delete DCItem if it was successfully added.

Otherwise code looks good, except that I would suggest add check that node actually has agent, because getInterfaceList will succeed if there are no agent, but node supports SNMP. But this depends on your setup, if you only have nodes with agents, then this is not necessary.

Best regards,
Victor

Hi!

Currently it's not possible. I have it in my to-do list. Depending on time, this feature will be in 1.2.1 or 1.2.2 release.

Best regards,
Victor

Hi!

I've fixed that. You can download updated source archive from https://www.netxms.org/download/rc/netxms-1.2.1-rc1.tar.gz. I also add two new options for configure:

--with-all-static

linker will produce really static executable, with system libraries statically linked too;

--with-static-subagents=list

specify list of subagents to be included into static agent. For example,

./configure --with-static-agent --with-static-subagents="logwatch ping"

will produce static agent with only logwatch and ping subagents.

Best regards,
Victor

Hi all!

I wish to update "Users" page on netxms.org web site. If someone is using NetXMS and wish/able to publish he's company name on that page, I'll be very appreciated. Ideal variant is to provide company name, web site URL and very short description of what is being monitored.

Best regards,
Victor

Hi!

I'm not sure I understand question correctly. If network is down and NetXMS server cannot reach mail server, then of course mail cannot be sent. For such situations you can use GSM modem attached to monitoring server and send SMS message instead.

Best regards,
Victor

Hi!

Currently the only working solution is to list all possible server's IP addresses in MasterServers. Using DNS name will not work because agent do name resolution only once at startup. I'll add feature request for supporting subnets in form xxx.xxx.xxx.xxx/nn in *Servers parameters.

Best regards,
Victor

Hi!

It's a new feature in development, which is not ready yet. Main idea is to add support for tabular data processing. It can be any data where instances are not known initially or can vary during time, and each instance has one or more attributes. Examples of such data could be process list or file systems list.

Currently Windows agent supports table FileSystem.Volumes. If you wish to take a look at it, you can create new table and enter FileSystem.Volumes as parameter name. You don't need to define columns manually. In last values, you will see new table DCI, and will be able to see it's last values. It is also possible to create charts based on table DCIs in dashboards.

Best regards,
Victor