Monitoring without worrying about the monitoring system
July 13, 2016
Monitoring without worrying about the monitoring system
Score 8 out of 10
Vetted Review
Verified User
Overall Satisfaction with LogicMonitor
LogicMonitor is our network monitoring system for our private cloud. It monitors all components in our infrastructure and alerts to a centralized network operations center that remediates the issue at hand. Consumers of LogicMonitor include our Cloud Operations team, Developers, QA, DevOps and Customer Support. The business problem it addresses is to try to catch customer impacting issues and remediate them before the customer even knows about them.
- Huge inventory of pre-built templates for monitoring things from Power PDUs to blade servers.
- Robust groovy scripting language that would enable administrators to add monitoring that is specific to your company.
- Uptime, uptime, uptime, uptime =). In the close to four years that we have used LogicMonitor, they have had one or two incidents of downtime.
- Eat to set up. The collector install takes minutes to install. I literally added hundreds of hosts within a day.
- Missing robust support for some fundamental things like SNMP traps.
- Support team is going through some growing pains. The people and coverage is growing but product expertise and customer experience is degrading. I am sure they will fix this though.
- They are in the middle of transitioning to a new UI. It looks better but is a lot harder to use for people that have been on the old UI for a long time. Probably just an issue with legacy customers.
- Money saved on capital expenditure to buy servers for a monitoring system. All our logicmonitor collectors are on virtual machines.
- Money saved on operating expense for headcount to maintain a huge monitoring system. Very low headcount to maintain a SaaS solution.
We had Zabbix before using Logicmonitor. We ran into massive scalability issues when trying to go past 2000 devices.
We spent a lot of time researching and trying to architect the system to work. However, despite our efforts we still missed alerts and customers were impacted due to the failed monitoring.
We had no scaling issues at all with Logicmonitor. We have tripled in size since we started using Logicmonitor and it all has been totally painless.
Using LogicMonitor
120 - Cloud Operations
- Network Administrators
- Network Operations Center
- System Administrators
- Database Administrators
- Telecom Administrators
Developers
Customer Support
- Network Administrators
- Network Operations Center
- System Administrators
- Database Administrators
- Telecom Administrators
Developers
Customer Support
1 - The minimum requirement is to know monitoring technologies in general. (e.g. SNMP, MySQL, http etc). Adding basic monitoring is a entry level position. Adding new monitoring will require scripting abilities or advanced knowledge of the product.
- Service uptime monitoring
- Capacity management
- Inventory
- Synthetic transaction monitoring (e.g. create interaction that simulates user interactions)
- Business intelligence reporting
- Amazon Cloud monitoring
LogicMonitor Implementation
- Implemented in-house
Yes - We started with a couple of hosts first to see how it works and if it would have any impact to our servers. Once we confirmed no impact, we migrated in batches of 20-100.
Change management was a big part of the implementation and was well-handled - None. Change management is processed already in case and we deployed Logicmonitor with that in mind.
- Silly mistakes done by administrators. (e.g. mistyped SNMP community string)
LogicMonitor Support
| Pros | Cons |
|---|---|
Kept well informed Support cares about my success Quick Initial Response | Slow Resolution Poor followup Problems left unsolved Need to explain problems multiple times |
No - There is no such thing with LogicMonitor. As a SaaS solution, support is backed into the cost of the product.
We added monitoring for RabbitMQ queue statistics. So the template would just poll the rabbitmq servers and get the stats for every single queue. After this monitoring was added, we noticed an increase in CPU utilization. Apparently polling 5000 queues from one rabbitmq server was taxing one cpu out of the server. When I brought this to support's attention, they were able to refactor the template and made it a lot lighter on the CPU. I thought that case showed excellent understanding and ability to solve problems.
Using LogicMonitor
| Pros | Cons |
|---|---|
Like to use Relatively simple Easy to use Technical support not required Well integrated Consistent Quick to learn Convenient Feel confident using Familiar | None |
- Adding hosts to the platform
- Adding groovy scripts for custom monitoring
- Adding collectors to expand capacity
- Clicking through the device tree structure when there are 20 host groups and 3000 nodes
Yes - I would say it is beta quality right now. The main webpage has single sign on but that does not work with the mobile interface. Mostly the mobile interface is designed to login to interface with current alerts.
Upgrading LogicMonitor
Yes - 90% of the time the release goes smoothly. Interaction with the customer can be greatly improved. (e.g. informing customer maintenance has been [done] and maintenance has ended). There was one one case where I found an issue after the release happened. There were a few cases where they found issues with the release and cancelled my rollout.
- Local cache of monitoring data. If network is down, the collector would cache the metrics for up to 30 minutes and will send it once the network is back up
- Website checks from collectors