Why Do You Export Data from LogicMonitor? We’d Love Your Input
Summary: Subomi Gbotosho from LogicMonitor initiated a discussion to understand how customers utilize the LogicMonitor API, focusing on reasons for data export and challenges faced. Several community members, including Jan Garaj and Andy_C, expressed exporting data due to missing functionalities and the need to visualize metrics better, often transferring data to platforms like SIEM, Excel, and PowerBI. Mitch Grande and Ruben Navas emphasized challenges with current reporting capabilities, leading to reliance on external tools for specific analytics needs. Members like Dominic Nguyen discussed the inadequacy of report functions and desired enhancements within LogicMonitor to avoid extensive API usage. Larsen Orchard highlighted the increased use of third-party MCP servers for better structured data access, pushing for LogicMonitor to potentially create a first-party solution.
Hi LogicMonitor Community
I’m reaching out from the LogicMonitor team as we’re doing some research into how customers are using the LogicMonitor API. We’re seeing significant usage of certain API endpoints, and we’d love to better understand:
- Why do you export data from LogicMonitor today?
- What kinds of data are you pulling from LogicMonitor?
- What tools or platforms are you sending that data to (e.g., dashboards, data warehouses, automation tools, SIEMs, CMDBs, other visualization tools etc.) and why?
- What would Logicmonitor need to do today for you to keep your data within our platform?
- The problems or workflows this API usage is helping you solve
If you’re comfortable sharing, even at a high level, your insights would be incredibly helpful and will directly inform how we prioritize improvements and enhancements to the API experience. There’s no expectation to share anything sensitive or proprietary, we’re simply looking to learn from how the API is being used in the real world.
Feel free to reply in-thread or message me directly if you prefer. Thanks in advance for helping us make LogicMonitor better for everyone!
Best regards,
Platform PM Team: Ted & Subomi
LogicMonitor
Larsen Orchard
·6 months agoHi LogicMonitor PM Team (Ted & Subomi),
Appreciate you reaching out and asking the community how we’re using the API. Reading the questions, I can’t help but suspect there’s a bit of a “question behind the question” here—namely:
Why is API usage climbing so fast lately?
And I think many of us can relate: API calls aren’t free on your side, and they’re harder to meter compared to feature-based licensing.
From what I’m seeing in the community (and inside my own org), a big driver is the recent surge in third‑party MCP servers that expose LM data safely through a structured model. These MCP servers have become incredibly compelling because:
A lot of companies (mine included) are pushing employees to use more AI,
…while simultaneously not approving additional budget for AI tools,
…which puts EdwinAI in an awkward spot—not because it’s bad (I hear very good things), but because it’s priced as a premium add‑on.
Meanwhile, an open‑source MCP server costs us nothing but some API calls. For many of us, that’s an easy choice.
The result is predictable:
More AI usage → more MCP usage → more LM API usage.
And honestly, the MCP ecosystem is becoming impressive. The quality of the open-source implementations is getting close to “product-level,” and they’re enabling workflows that would otherwise require paying for EdwinAI or building internal tooling.
I’d personally love to see LogicMonitor release a first‑party, supported MCP server someday. It would help standardize the data models, reduce unnecessary API traffic, and keep more of the value—and the conversations—within the LM ecosystem.
Until then, many of us will keep leaning on those community-driven MCP projects. They’re powerful, they’re flexible, and—let’s be honest—they’re free at a time when budgets aren’t.
Jan Garaj
·6 months agoI would say the problem is limited LM API (LM query language). Let's say my AI agent needs to find high CPU usage (not based on alerts, but based on CPU metrics) across 1k resources. It must to execute at least 1K MCP/API requests with current LM API implementation (I would say at least 2k+ actually). There is no official LM API endpoint, which can be used to execute single LMQL across 1K resources. This is typical use case how LM API doesn't scale. Add additional signals (logs, traces) into picture and requirements on the LM API will explode.
Dominic Nguyen
·6 months agoHi Logic Monitor team,
I'm glad I found this community. I have been trying to create a dashboard of all our WAN link in a dashboard for our executives MOR (Monthly Operating Review). Our Executive would like to see an indicator of all our WAN links and which one has potential bottleneck.
In our case, we'd like to report the 95th Percentile of these WAN links in this format:
Location Device Interface Bandwidth(Mbps) In_P95% Out_P95% In(Mbps) Out(Mbps)
Location = a custom property that we sync from CMDB to each device in Logic Monitor
Device = Device Name
Interface = WAN Interface (we already have a way to identify all WAN interfaces in LM)
There are # of issues with the current Logic Monitor reports/dashboards today:
The metrics value is different between Cisco devices vs Viptela devices
There isn't any dashboard or report that can provide all the data in the table. We have to combine multiple reports/dashobards/API scripts to extract the right data we need
The biggest issue is that we found the P95 calculation in Logic Monitor is either misleading or not correct (I did the math and it is way off, but I am not sure how LM calculate this data)
Therefore, we are thinking that if we can export the raw data for the WAN interfaces then we can calculate the proper P95 value per day and build our dashboard to show the heat map on how each WAN interfaces perform day to day.
By the way, we'd like to have an indicator if the P95 > 80% (Red-Critical), 60% < P95 < 80% (Yellow-Warning), P95 < 60% (Green-Normal).
To answer your specific questions:
Why do you export data from LogicMonitor today?
Because there is limitation of Dashboard/Reports in Logic Monitor for what we need
What kinds of data are you pulling from LogicMonitor?
We want the raw Utilization for In/Out, Bandwidth, device custom property (location)
What tools or platform are you sending that data to ...
We're planning on sending it into PowerBI for all the data calculation and dashboard/report
What would Logicmonitor need to do today for you to keep your data within our platform?
I'd love to see that LogicMonitor enhancement to provide they way for us to generate report or dashboard within LogicMonitor, then we don't have to export the data to another tool. Or this there is a way for Edwin AI to injest the data and provide the dashboard/report there, that would work too.
The problems or workflows this API usage is helping you solve.
The API usage is very cumbersome and take too long to export all 400+ WAN links to run.
I hope this information is helpful and I'll monitor this thread to see if there is any development on this. If you have any question or need me to elaborate the use case, I'd happy to do so. Thanks!
Dominic
Dave Lee
·7 months agoWhy do you export data from LogicMonitor today?
Reporting, billing, CMDB integration
What kinds of data are you pulling from LogicMonitor?
Alerts, metrics and inventory/device properties
What tools or platforms are you sending that data to (e.g., dashboards, data warehouses, automation tools, SIEMs, CMDBs, other visualization tools etc.) and why?
data warehouses (which drive dashboards on our customer portal - we want the data on our own platform so we can display it to customers in a format consistent with other data we present, and to prevent the need for them to log into LogicMonitor)
automation tools
we have an incident enrichment tool that looks at device properties
we have a nightly process that queries vendor support APIs and pushes data into LM devices as properties. We didn't want to put this load onto customer collectors (which often aren't allowed to query public IPs) and this allowed us to use custom properties, rather than auto properties.
processes to automate customer billing based on categories of monitored devices
we're likely going to build some automation to ensure servers/switches in monitoring are defined in our PAM/remote access solution
Updates of Service Now ITSM incidents outside of the capabilities of the native integration. We have some batch processes that re-open incidents in some situations (e.g. closed Incident related to open alert.
LM -> Service Now CMDB integration (soon - I hope!)
What would Logicmonitor need to do today for you to keep your data within our platform?
We are considering looking at a solution to export more data from LogicMonitor into a data warehouse (potentially PowerBI) so we can provide our internal users with a better reporting solution. If the solution in LM were more flexible, we could perhaps use it instead.
The problems or workflows this API usage is helping you solve
Quite a few listed above. Another we are considering... building modules for each device type to check that the monitoring is working correctly. Similar to the SQL Troubleshooter module, but something that would make API queries to check that the proper modules are applying to a device. Currently, if a module doesn't apply, then there's no method to highlight that something in monitoring isn't working.
Subomi Gbotosho
OP6 months agoThanks so much for this Dave, could you please help explain further your expectation if how the "reporting solution" would work?
Ruben Navas
·7 months agoWhy do you export data from LogicMonitor today?
To import a set of polled data to a 3rd party database used to implement dashboards that combine different sources of data (inventory tools, monitoring systems, etc.).
What kinds of data are you pulling from LogicMonitor?
Usage of connectivity resources like WAN links.
What tools or platforms are you sending that data to (e.g., dashboards, data warehouses, automation tools, SIEMs, CMDBs, other visualization tools etc.) and why?
Other visualization tools that support the capabilities demanded by our stakeholders that are not available in LM dashboards and reports.
Filters can be enabled in dashboards so that the users can reduce the amount of data displayed on the dashboard. As dashboard data cannot be exported to a CSV file, we can use reports. However, reports do not support filters.
What would Logicmonitor need to do today for you to keep your data within our platform?
Improve the existing dashboards and reports. Missing capabilities & features have been identified and shared with our LM representatives. So far, most of them are still not available.
The problems or workflows this API usage is helping you solve.
Export data in an efficient way. Available API endpoints can be used to export a huge amount of data without filtering that must be filtered before processed. Data should be queried from LM using resource/instance filters in order to reduce the impact on the LM side.
Subomi Gbotosho
OP6 months agoThanks so much for this Ruben, in you response here: "Other visualization tools that support the capabilities demanded by our stakeholders that are not available in LM dashboards and reports."
What do you mean by capabilities, I can raise these again and bring them to the team. Thanks!
Mitch Grande
·7 months agoAndy_C
·7 months agoIn short we need to provide basic graphed metrics to non technical people about the state of their systems. Frequently its just CPU / load average , memory and disk space % used. We "customize" the layouts to our / their corporate branding standards and presented by their CSM / TDM etc.
Generally these are for general distribution inside the customers premises and are kept for historical reasons. We have no intention or desire to give end users full access to the portal, even for their own gear.
Jan Garaj
·8 months agoBecause missing functionality, problematic UI/UX in the LM
Audit logs, metrics.
other visualization tools, because missing functionality, problematic UI/UX in the LM
I would say this is a problem in the logic. There always be something better for customer use case. But LM is always trying to keep you ONLY in the LM = vendor lock in. Typical problem is metrics - I'm happy to have metrics in the LM, but then I want to have own "Excel" for visualization. So I point my Excel to API (so it is just querying, not exporting) and then you realize how complicated is metric data model (all ids, which must be in the query just to get CPU of server X), then you realize then API has limit how many datapoints you can return - so there is no way to visualize for example month of CPU usage in the Excel. Then you will scale it on the enterprise level and you will realize how low API rate limits are (BTW that's applicable also for Catchpoint API).
No real timeseries/analytics on top of audit logs. Missing widget types and missing widget customization - can't create own widget type, so I have to stick with out of the box widgets only (yeah, LM support will say: you can write own JS in the HTML, but I will say NO, because security), dark mode, ....
Subomi Gbotosho
OP6 months agoThank you for this Jan, do always appreciate your insights. Definitely helps us prioritize on what we need to work on