struggling to configure a packet loss alarm

[brettjones]

Hi there can anyone assist me?
bit of a newbee at netxms - just moving from solarwinds to this and i am finding creating my own alarms a bit difficult.

i got about 250 nodes that i need to monitor high packet loss on (IE something that says if packet loss is greater than 30% then send an email and alarm)

but i am struggling on how to do this.
Looking at some of the other posts i found this from Victor:

"2. Load PING subagent on any managed node, and use any of ICMP.xxx parameters to send ICMP pings. In this case, you can monitor not only reachable/unreachable status, but also ping response times. You can find additional information about PING subagent here: https://www.netxms.org/documentation/ping_subagent.shtml."

Best regards,
Victor"

but how do i load PING subagent onto the managed node?

I have checked the documentation but i dont get how to create this ping subagent.
then once i have done this how do i create an event in the event config and event process that will trigger an alarm when this happens and send me the mail?

any help anyone can give me on this i would be eternally grateful!

thanks all

Brett Jones

[zev]

Hi.

There is information about ping subagent:
https://www.netxms.org/documentation/adminguide/subagents.html#ping

It is loaded with "SubAgent = ping.nsm" line added into agent configuration file. There is some information about agent configuration: https://www.netxms.org/documentation/adminguide/agent-management.html

From the next release ping response times can be collected without subagent: https://www.radensolutions.com/chiliproject/issues/511

About sending email notifications can be found there: http://wiki.netxms.org/wiki/How_to_send_e-mail_notification_when_disk_space_is_low

[brettjones]

hi, thanks for this but how do you apply the subagent to the event configuration so it will alarm and send a mail?  there is nothing to tell us how to actually do this.

thanks

[zev]

Hi,

Subagent just adds functionality to agent. So when subagent is loaded to agent it usually just add some additional DCI's.

So you load ping subagent, go to Data Collection Configuration, add new parameter. Origin of parameter should be NetXMS Agent, click Select button, select "Icmp.PacketLoss(*)", and replace * with IP adress. Now configure threshold  as shown in this(http://wiki.netxms.org/wiki/How_to_send_e-mail_notification_when_disk_space_is_low) instruction. In your case event should be created if value is more than 30. You can create your own events that will be generated once rule is violateв and other when it is back to normal.

Afterwards same error code will be used in event configuration.

Did this answer your question? If not what exactly step is confusing?

[brettjones]

hi there, thanks for your help so far.
we definitely now have a better understanding... however we are struggling to find the parameter "Icmp.PacketLoss(*)"

we have no parameters under netxms agent.
all the config file looks fine and auto agent is on etc.... so where do we find the icmp packet loss agent from?

kindest regards

brett

[zev]

Please check that when subagent is loaded there is message in log:
[19-Sep-2014 18:58:35.863] [INFO ] Subagent "ping.nsm" loaded successfully

Ty to do manually configuration poll for this node.

Check that origin of DCI is set to "NetXMS Agent" before you click "Select..." button.

[brettjones]

hi there!      still struggling to get this working!!!!!!

ok -

got the subagent loaded.

followed wicki and instead of doing it on individual nodes (i have over 200)  i have created a template following the guides you sent.

question.  if this is a high packet loss alarm then what instance should i add?  icmp.packetloss(*)    should * be the end point ip address of the CPE?
if so then i guess the template wont work as ill have to add the ip address of each router individually.
on another note on my experimental router i have changed the 8 to the b end cpe ip address..  no luck it dosnt work.  it just tells me first the node is down....
then tells me the interface is down....
but no packet loss alarm!  :-[
ive even set the poll times so that packet loss is every like 5 seconds.  but no it fails to work...... 

help?  what on earth am i doing wrong????????

[brettjones]

hey all,

can anyone assist?

still cant get this working.

kindest regards