NetXMS Support Forum

Quick one...

How would you handle regex in the DCI transformation?

Let's say you poll an SNMP OID which returns a string similar to "MyVal=18#-99#AB_01".
You want to extract the "-99" and return an integer to graph it...

In Perl it would sth like $1 ~=/#(.+)#/ - if I'm right.

Regards,

Jdamien

Hi there,

Some things that could be interesting based on experience:

- When creating a node, reverse lookup would be interesting. Sometimes one knows the IP but not the FQDN. If you enter an IP @ and push "resolve" you get IP @ of localhost (the one running the console) and no name.

- when defining a new event processing rule, we hav eto close the window to get asked if we want to save the story. Would be nice to have a drop down menu with options (like duplicate rule, etc...) among which "save" to save the current config if relevant (if no change then greyed out ;-)

Regards,

Jdamien

Hi there,

Some suggestions...

I noticed that when an event (normal) clears out an alarm (critical) and you request an email to be sent with a message built up using macros, well the email is nearly empty as the macros are returning empty strings (%M, %A... only severity and time/date are given).
Email subject is: [%S] %A - only severity returned
Body content is: Alarm: [%S] %A\nNode: %n [%a]\nTime: %t - %A is not returned!

Using %m is not always convenient!

All this because the cleaning event is not creating any alarm. The only flexibility is about the message key for cleaning related events.

It would be nice to still be able to define a message text available for action such as sendmail.
You will most probably suggest to use %M or %[name] but for some scenarios no script is used. So it is quite cumbersome to have to define a script just for the sake of having a text message available.

Suggestion: get events configured as for alarms and give the option to display event in alarm list or not. Then message text still exists and is available for actions. Further extension: active event list, history event list, archives. Created events are displayed in alarm list. When ack'ed or terminated, then instead of trashing them from the alarm list, park them in a history event list (new window) for a while (configurable) then trash it... or preferably archive it in a text file.



Best regards,

Jdamien

Aarrggg!

I did not notice the new menu "Map" coming up when in the scope of "Map".
Many thanks for this pointer.

Best regards,
Jdamien

Hi there,

when defining a data collection, it seems that values are not accessible after a while (how long? I don't know).
One thing for sure is that for values polled every 30 sec it is not possible to draw a graph for the last month and after 2-3 months you cannot even access any data. It seems as well that the values are still updated but you cannot access any value even the most recent one!

I guess there is sth wrong here. right?

regards,

JDamien

Hi there,

When looking at the map (Default or any you may define), it looks like if you arragne your icons and save it... your job is gone next time you visit it - even after having saved it (via icon or answering "yes" at question when closing window).

Is there sth I forgot or is this a bug?

You keep the background image but all the icons are stacked on each otehr at the upper left corner again.

Best regards,

Jdamien

Hi,

I wish you all the best in this job 
Rgds,

Jdamien

Hi

One more in this topic...

Using $event->message we get the message text defined in the "event" panel.
How do you differentiate between this message text and the one you can re-define in the "event processing" panel when defining an alarm? Is there a property to the $event object that covers this one?

In the script, the $event->message refers to the text message defined in the event not in the alarm which can be more specific as we get additional macro's at that level, right?

Regards,
Jdamien

Hi again,

Seems I managed to crash my server while returning an array...
Is the following supported?

- Action which calls an action on remote agent having as parameter the output of %[script] (I only have '$1' as parameter to agent script in agent config file)
- Output of %[script] is an array of 2 elements
- Immediate effect: console connection broken and server process not running anymore!

Nothing in server log!
No trace of EXEC request for action related script on agent in agent log!

Regards,
JDamien

Great - wonderful 
Your answers just clean off all my questions
many thanks for this comprehensive feedback


Best regards,
JDamien

Dear all,

Some generic questions related to script definition:
- Any possible way to insert comments? Using for example # as in Perl...
- Is the operator ~= similar to the one in Perl? I.e. calling for regular expression?

As far as reg.ex. are concerned, is the syntax the same?
If for example, I need a script parsing a message text to return different elements and use them to defined return argument(s)...
Message test is "TCP Socket is open on host xyz.domain.com (172.16.0.1) using port 3333" and my regular expression is:
$event->message ~= "^TCP Socket is (.*) on host (.*) \((.*)\) using port (\d)$"
How can I put the matched info in different variables: $status, $host, $ipaddr and $port?
How do you define those variables?

What about functions in script?

How do we return an array as argument?
Is syntax return ($val_1, $val_2); correct?
Will this array be returned as an array?

Would you have a complex example to post that would illustrate as many things as possible about defining NXSL script?

Regards,
JDamien

Thanks for this rapid feedback!

Regards,
Jdamien

Jesus... that's great!
Thanks for the info...

But you know my next question, don't you ;-)

Where can I find a description of $node and $event variables/objects?

So I found $nodes object attributes:
Attributes of node object:

name      Node name
id         Object identifier
status      Node status (0 = Normal, 1 = Warning, 2 = Minor, 3 = Major, 4 = Critical, 5 = Unknown, 6 = Unmanaged)
ipAddr      Node primary IP address (as text string, like 10.2.1.4)
isAgent      Agent presense flag: 1 if NetXMS agent present on node, 0 if not
isSNMP      SNMP support flag: 1 if SNMP is supported by node, 0 if not
isBridge      Bridge flag: 1 if node is bridge or switch, 0 if not
isRouter      IP router flag: 1 if node is Ip router (can forward IP packets), 0 if not
isPrinter      Printer flag: 1 if node detected to be a network printer, 0 if not
isCDP      CDP (Cisco Discovery Protocol) support flag: 1 node supports CDP, 0 if not
isSONMP      SONMP (SynOptics Network Management Protocol) support flag: 1 node supports SONMP, 0 if not
isLLDP      LLDP (Link Layer Discovery Protocol) support flag: 1 node supports LLDP, 0 if not
snmpVersion   SNMP version used for communication with node's SNMP agent
snmpOID   SNMP   object ID
agentVersion   NetXMS agent version (as string, like 0.2.26)
platformName   Platform name, as returned by NetXMS agent (for example, windows-i386)

Remains $event object attribute...

Regards,
JDamien

I would appreciate if someone could take some time to put some inputs here on this subject...

Possible? Not possible?
Another example: Imagine you want to send an email to different staff in function of the alarm/event... do you have to create a specific action for each email?

Thanks in advance for ur support and time.
Rgds,

Jdamien

In the same context...

Would be nice to pass as argument the email addresses for the "sendmail" action!
And why not also content and subject or a specific template... well a bit more flexibility around here would be very powerful.

Cheers,

Jdamien