LogicMonitor Review

The real strength of this product for me is the infrastructure device support.

What is most valuable?

Integration with a wide variety of infrastructure systems is the most valuable feature; most critical to us was VMware integration and Citrix NetScaler integration. We use a lot of other products with LM, as well, but the biggest gaps we had with trending data were load balancers and VMware. (vCenter itself has some capability, but is quite limited by comparison.)

A close second is the dynamic "top 10" graphs. You can have the system look at up to 520 data points and just show the top 10; we use this all over the place. If I have 100 VMs and want to see CPU usage or Disk I/O across them, that is a lot of lines in a graph (too many). But with the dynamic top 10, I just see the most active systems, which 99% of the time is what I care about. The system automatically adapts to adds and removals of things (with wildcards in the custom graph definitions), so maintenance is very low.

It is also not difficult to write custom things for LM; myself the biggest thing I adapted to LM is HPE 3PAR monitoring, something I had been maintaining on the side for many years now with Cacti. With LM, I get much more data, and it's 1000 times easier to manage than Cacti. I collect several thousand data points per minute from our 3PAR arrays.

We have not used their alerting, reporting, or service-checking features of the product. All of the value for us comes from the graphs and dashboards.

Monitoring something like a Linux system is easy, but pulling out thousands of metrics from vCenter, or NetScaler, and making dynamic graphs when things are added or changed is an enormous time saver.

LogicMonitor is by no means the only monitoring product we use; we use many, but it does things none of our other tools can hope to do, so it is critical to monitoring our services.

How has it helped my organization?

Good, accurate dashboards: I can pull data from multiple sources, whether it is VMware CPU Ready stats, server CPU utilization, application metrics (MySQL, etc.), load balancer metrics, and even 3PAR storage metrics and show them all in one screen. I have seen many products that claim to do this kind of thing, but LogicMonitor is the only one I have personally seen/used that makes it easy to use.

What needs improvement?

I much prefer the older UI for managing devices and dashboards. They gave assurances that they would not retire the old UI before making it feature parity with the new UI, but they did not follow through with that.

I was a VERY happy customer of LM up until a couple of weeks ago, when I noticed they flipped the bits to disable the old UI. That made me very, very upset.

For how long have I used the solution?

I have used it for about two years.

What do I think about the stability of the solution?

We have had problems with the Windows agents crashing on occasion and not automatically restarting; after engaging support on the issue, no resolution was found, so I just have to manually restart them when they go down (at this point, it seems like once every few months). No issues with stability on Linux agents.

What do I think about the scalability of the solution?

We have had to deploy a larger number of agents to support our devices than I expected, 16 agents for 250 devices currently being monitored. We had some issues with high CPU usage on agents earlier on; filed some support tickets and got a few issues resolved. For the most part, things are fine now; we just have to have a bunch of agents.

How are customer service and technical support?

Technical support is pretty good; I have been happy with most of my tickets being resolved.

Which solution did I use previously and why did I switch?

I have been doing custom monitoring things since 1997 with MRTG, then custom RRDTool monitoring, then moved to Cacti, now LM.

How was the initial setup?

For what we wanted, initial setup was very straightforward; it was one of the easiest monitoring systems I have ever used. The caveat is, we are not using alerting or reporting, so maybe there is more complexity in there. But for basic graphing and dashboards with custom graphs, it's amazingly simple, and I can do many things that simply were not possible with previous solutions that I had been using.

What's my experience with pricing, setup cost, and licensing?

The per-device licensing is painful for virtual machines. (Virtual machines make the overhead of segmenting things out more much simpler, but per-device licensing like this hurts that model.) The per-device licensing model is common among this type of app, whether it is LM, or something else like New Relic, or even Chef, which we also use. The value proposition isn't there for us to deploy LM across all 900+ devices/VMs that we have. So we target it at the most high-value areas and use other tools.

Which other solutions did I evaluate?

I had been in depth with Cacti for several years; wrote a ton of custom code to integrate with systems and network devices. Before that, RRDTool. I had looked at various other systems like Zabbix, and other names I don't remember at this point. Many years ago, one company I was at used Mercury SiteScope; that product at the time was still limited, as well. A full three years after they deployed it (after I left the company), they were still using my custom graphs to do things that the very expensive SiteScope could not do.

The real thing that got LM in the front door for us was visibility into VMware. Specifically, I was quite concerned about monitoring CPU READY states; this metric is only available on a per-VM basis in vCenter. I can't go to a VMware host or cluster view and say show me CPU READY. It doesn't work like that. LM shows me this information very easily. Many tools out there do basic SNMP support, but don't have more in-depth knowledge about what each of those things actually mean. I have been using SNMP for almost 20 years and I still consider it to be a black art. LogicMonitor doesn't support every device in the world, but for those it does from a SNMP perspective, it knows what the data points mean and creates accurate graphs to represent that data. Talking to APIs in VMware to get information is very useful as well; it's not something that is easy to integrate with (relative to other products, because you have to know the APIs and understand the responses).

What other advice do I have?

The real strength of this product for me is the infrastructure device support (https://www.logicmonitor.com/technologies/). LM is a great tool if you have a lot of things and want to be able to create dashboards to get status into many devices at once without a lot of work and without vendor-specific tools (going to one app for network stuff, another app for storage stuff, another app for virtualization stuff). If you use products that are NOT on their list, it is still a good tool; it just takes some more work to integrate. I have written custom integrations for Varnish and Memcache (overriding their default integrations, which were not adequate for our needs); it is not complicated.

Disclosure: I am a real user, and this review is based on my own experience and opinions.
1 visitor found this review helpful
Add a Comment
Sign Up with Email