Shack
AdvisorJoined 9 years ago
User Profile
Advisor
User Widgets
Contributions
Meraki Security Appliance Tunnels
We monitor several large Meraki networks for customers, and we’ve recently switched to LogicMonitor’s newer Meraki monitoring method for a few of them. One challenge we're running into is with the Meraki Security Appliance Tunnel datasource. With the Hub(Mesh) configuration, each Meraki security appliance gets a tunnel datasource instance for both WAN1 and WAN2, even if only one WAN is in use. The issue: when a firewall at Location A goes down, it triggers an alert from the Cisco_Meraki_SecurityApplianceHealth datasource (which is expected). But in addition to that, we also receive multiple alerts: A tunnel alert for every tunnel on the device 2 tunnel alerts on each remote firewall in the mesh that has a tunnel to that location(wan1 & wan2) This results in a flood of alerts, even though it all stems from a single tunnel being offline. To cut down on the noise, I tried building a Service and adding all the tunnel instances to it. That helps, but the Service alert only says the service is out of compliance—it doesn’t tell me which tunnel is down. Has anyone come up with a better way to monitor Meraki tunnels using the newer integration, without generating excessive alerts when one device goes offline? I'd almost prefer some ORG level device like the legacy method has and put the tunnel connections there whether they are Site-to-Site or 3rd Party. When a tunnel goes down I don't need every FW in the environment alerting that the tunnel is down.Cisco Wireless Access Points/Wireless Lan Controller
It's been recently brought to my attention that if an access point gets renamed that's being managed by a 9800 series WLC it will show up as a new instance the next time active discovery runs instead of updating the existing instance. I believe this happens due to how SNMP handles the rename and a new instance is created in the OID index but I have yet to confirm other than what I see in my portal. Now you'll have 2 instances with the same MAC and SN but different names both reporting statuses on your controller. If the AP goes down you'll get 2 alerts. Outside of it being a manual QA process I'm wondering if anyone else experiences this or has a good suggestion on how to detect duplicates and/or a workaround? We already had to modify our AP datasource so that AP's are not removed during Active Discovery because when they go down they fall off the OID tree and that was deleting our instances and clearing alarms.Re: Complex datapoint - groovy script to calculate the value from the device property
You can't use ## in groovy. I think those are reserved for PowerShell. Try something like this. You can do hostProps.get or instanceProps.get depending on where the property is. def answer = instanceProps.get("auto.myproperty123"); if(answer == "true") { return "0"; } if(answer == "false") { return "1"; }
Groups
Inner Circle
Welcome to the Customer Showcase, sponsored by “The Inner Circle “. Feel free to peruse this Customer Story Showcase to read more about what our customers are doing out in the world and how LogicMonitor is supporting their journeys.
