New UI Impact Series - Topology Node Grouping
Next up in our series is Topology Node Grouping. This new feature allows users to dynamically group nodes in saved topology maps based on up to three levels of property metadata. By leveraging tags and labels stored as LogicMonitor properties, you can now organize your complex network maps into intuitive, property-based clusters.The groups are automatically color-coded according to alert status, providing an instant visual indicator of potential issues within specific node groups. So, how does this help you troubleshoot more efficiently? In complex network environments, identifying the severity level and location of issues can be like finding a needle in a haystack. Topology Node Grouping transforms this process, allowing you to quickly assess the 'blast radius' of any network problem. For instance, in a map of virtual machines, grouping by location could instantly reveal that all alerts originate from a specific data center. This level of clarity, which would have required extensive zooming and manual inspection in the past, is now available at a glance. By speeding up the identification of affected areas, Topology Node Grouping enables IT professionals to respond more swiftly and effectively to network issues, potentially reducing downtime and improving overall network performance. Want to know more about Topology Node Grouping? Check out these articles on Node Grouping and Searching for Nodes.
Needed Features for Network Topology Mapping
We need the ability to create topology map of our switches and routers showing the actual links between interfaces on the devices and showing the latest InMbps / OutMbps datapoints for each interface. The goal is to have a live network map for networks likeInternet Service Provider networks, to be able to visualize the active state of the network and the bandwidth utilization and alert status of each device and interface / link. We need to be able to do the following: - Manually add the switch and router resources to the map. - Once the resource is on the map, click on the resource on the map and select an interface instance from the "network interfaces" LogicModule to add to a "Link". Then drag over to, or select, the connected resource, and select the associated interface instance from the "network interfaces" LogicModule on the second device. - This will visually show a link between the two devices, and the InMbps and/or OutMbps datapoints should be displayed over the link on each end of the link. - The color of each half or end of the link should also change based on the highest active alert level for the associated "network interfaces" instance on that end of the link. So for example, If we have Router1 interface GigEth0/1 connected to Router2 GigEth0/2, and Router1 GigEth0/1 has an active warning level alert for InBroadcastPkts, then that half of the line representing the link, should turn to yellow. - It's not enough to simply outline the resource with a color based on the alert level of the resource. We need to visually see each desired interface / link on the map and see visually which interface(s) is/are alerting. LogicMonitor has done an absolutely outstanding job of giving us immense flexibilty in how we can represent any datapoint on widgets on dashboards. But when it comes to the Topology mapping, we’re severely limited in how we can represent the exact same datapoints visually on the map. Also, for these kind of use-cases, people managing these kinds of network generally already know how their devices are connected, or can easily identify the connectivity. We don’t really need LogicMonitor to automatically detect connections between devices, because that’s too big of a challenge to automate that well. We just need to be able to manually say, “This interface on this device connects to this interface on this device. That’s a link.” The link itself is actually comprised of two sides, each being an instance on theresource’s LogicModule. And it could be “Interfaces 64 bit” or “Network Interfaces” or whatever other LogicModule we need. When we make w widget, we can select whatever LogicModule and Instance we need. We should be able to do the same thing on the topology map.API: How do I bulk-query topology data?
I need to query topology data for all devices. What is the endpoint for that? Currently, I’m using: topology/data?resource=device.XXX&algorithm=I-Feel-Lucky ...but obviously, this requires iterating over all devices, which is stupidly inefficient and is clogging up the REST API. How can this be achieved in a single (albeit probably paged) query?
Topology Mapping in 1.85
Good Morning All. So while my organizations had some headaches with Topology Mapping I’m curious if anyone is having this issue: Basically when creating a Topology Map, the entire map will reset every 5 minutes within the editor (Working on the Dynamic Mapping). This is a bit difficult to deal with and hasn’t been an issue in 1.84 (granted our issue was the map would reset itself after saving to default, but we could at least get that far.) If anyone has any issues similar to this or has any solutions to prevent this, I’d be happy to hear them. This is the first time in months that we cannot make an accurate map of our core (basically showing our primary back-bone topology).
A DataSource to Troubleshoot ERI Merging
One of the most common behaviours noticed in topology maps is ERI Merging. This is caused when two or more devices share the same identifier (ERI). The example I always like to give when I’m teaching Topology Mapping is the word “football”. To a European like myselfthis is a game played with your feet, however in other parts of the worldthis is an altogether different game. Now let’s imagine we have a Topology Map connecting various sports together; what would show up if the map connects “football” to “basketball” - would it be the kicking game or the throwing game? Well, in LogicMonitor, it would be effectively indeterministic to tell. The two games would merge into a single object in the map (they merge into one of the resources at “random”). A key indicator of merging is one device showing as another device in the topology map. Luckily, there are a few out of the boxways to overcome this merging - the topo.blacklistand topo.namespace properties. If you’re interested in finding out more about merging how these are used, I have created a LearningByte which you can watch for free in LM Academy here - you will need a free Academy account created first. https://academy.logicmonitor.com/topology-mapping-toponamespace-topoblacklist/1329206 In order to use the blacklist property, you must know which ERIs are being merged. This can be discovered in the UI through a manual comparison of ERIs between resources (you can export to excel and process there if you’d like), however this can be a cumbersome process and doesn’t reveal how many resources are merged. That’s where my new ERIMergeTroubleshooter comes in. Using the LogicMonitor API to run the !erimergelist and !erimergedetail collector debug commands, it creates one instance for each merged ERI and a subsequent instance level property listing which other resources merge with that ERI. For example, we can see that this “Router” resource has merged with a “Server” resource. Applying the troubleshooter DataSource, it immediately reveals which particular ERI has merged, and which resource it has merged with (this is a trivial example, although most situations are often more complex). If you’d like to try out this custom Logicmodule, it can be imported now from the LMExchange (locator: F26PEJ); it will be great to hear some feedback from real world testing! Caveats: By default, this applies to all resources in the portal, so users should modify the appliesto if they require testing on specific devices only The module has not been tested against or developed for chained ERI merging API credentials are to be added as device properties based on the technical notes Thanks!
