What is up with Meraki API Modules?

Two netscan improvements are coming your way soon.

1.  In portal version 220, if the Meraki Enhanced Script Netscan reports a different IP Address than is in system.hostname, it will automatically update the system.hostname and system.ips values on the associated devices in your portal. Currently, we rely on the IP address and device name, but this changes to using the meraki.serial as the device's unique identifier.

2.  In Q2 (April-June), we're planning to implement a "test" button in the Enhanced Script Netscan, so you can see what it would do if you ran it.

Honored to have the legendary Patrick_Rouse comment on this post. :) And of course, I have tried multiple support tickets on these issues. More issues just keep popping up the more I work with the API integration. I don't believe I've made any modifications to the LM Meraki API modules in our portal, so that shouldn't be causing issues. Interesting you mention meraki.snmp or SNMP at all, for that matter, as I thought the API calls were supposed to essentially replace the need for SNMP. Still, we have SNMP properties on our devices as well so I don't think this is causing any issues.

Patrick_Rouse wrote:

For the modules to work, you "need" the following properties on each device (set directly or inherited)

To this, I would say this is by far the most tedious thing as the Netscan almost never pulls in all the right properties by default. Furthermore, I believe my individual issue of thousands of Warning Alerts (it jumped up to 4500+ overnight) was also due to the inheritance you mention and it not working as expected. I'm still working with support on these issues as I have no idea why inheritance doesn't seem to be translating down to the devices in this regard. Also weird that only some devices were alerting, and not all. If it were truly an issue with one of those properties you mentioned above, I would expect all devices in that Org or Network would be alerting--especially because I set the meraki.api.key at the parent level and meraki.api.network at each site folder underneath that. But I'm working on that with support right now.

Just mostly meant to say that this stuff is so tedious and almost feels like more work than it's worth sometimes. But I'm not giving up on it yet. I agree with Allister that a Wizard of sorts would be nice. I mentioned in one of my support tickets for Netscan that it would be nice to see what kind of changes to expect before "approving" them as I've had the Netscan remove devices in my portal with no warning whatsoever. That Wizard or confirmation screen would be nice.

I'm sorry you're struggling, but we'd be happy to help.  Some key points.

If something is affecting your ability to monitor, please contact LogicMonitor support.  However, if it's not urgent or you're just looking to learn from others this is a great place to do that.

The Cisco_Meraki_API DataSource is the module that collectors use to get all of the data from the Cisco Meraki Dashboard API.  I'd advise that you not edit this module or change the polling frequency, as this module has internal timing that it uses to gather data that the other Meraki modules rely on.  The current version of this DataSource is 1.31.0 and no longer uses any per-device API calls for Cisco Meraki Switches, which was a limiting factor for larger customers with thousands of switches.

If you're getting no data for the client count in any of the performance modules, it'd because the meraki.snmp settings are invalid or your collector cannot access snmp.meraki.com on port 16100.  That Meraki Cloud SNMP service only provides LM with the client count for the performance modules and send/receive data for the Meraki Interfaces modules (not Switch Interfaces).

For the modules to work, you "need" the following properties on each device (set directly or inherited)

• meraki.serial (this is device specific)
• meraki.api.org (this is the organization ID for your Meraki org)
• meraki.api.network (this is the network ID for the Meraki devices at a given location)
• system.categories (if you don't add a value here, the PropertySource  "addCategory_Cisco_Meraki_Device" will set this for you "when it runs").  The NetScan adds this for you, but if you add devices another way you can either let the PropertySource set this or you can set it manually based on the device type.
  
  The Resources Tab here has a PDF that shows what the Meraki Monitoring for each device type should look like -> Cisco Networking App Marketplace
  
  
  Screenshot from a device in my portal with the properties set.
  
  

  
  Resources:
  
  Cisco Meraki Monitoring product documentation

  Cisco Networking App Marketplace
  
  Create a LogicMonitor Support Case
  
  
  

I have to agree. Also the Netscan is clunky & error prone. It needs to be 'Wizardized' so that you just enter the details you have and any optional filters and it takes care of the rest.

Also, the error handling needs to be improved. I remember one time spending an hour trying to figure out why some devices wouldn't add when it was due to an empty folder sitting in the recently deleted items which wouldn't allow the same folder name to be re-added, but the error messages & debug doesn't point you in the right direction...