# Server Monitoring

OpManager's server monitoring features help you ensure ideal uptime, performance, and fault resilience for your Windows, Linux, Virtual, Exchange, Database, and Domain controller devices

- Track over 300 server performance metrics
- Monitor 130+ server vendors out of the box
- Monitor Windows, Linux, Solaris, AIX, & UX servers
- Monitor & manage critical processes & services
- Prevent server hardware failures with instant alerts
- Automate remediation and troubleshooting

![Hardware Monitor Application](https://cdn.manageengine.com/network-monitoring/sem/images/server-monitoring/opm-server-monitoring-video-thumbnail.png)

## Key server monitoring capabilities supported by OpManager

OpManager features agent-based and agent-less mechanisms to monitor server devices. After scanning and discovering your devices, OpManager identifies your servers and classifies them based on various factors like vendor, type, and functionality. You can [monitor server uptime](https://www.manageengine.com/network-monitoring/how-to-check-server-uptime.html?server-monitoring), processes, services, as well as key metrics like the uptime, CPU, memory, processes, services, and disk space. Let's take a look at each of these capabilities.

### Ensure 24/7 uptime with server availability monitoring

**Ensure high-availability for your servers with regular uptime checks.** OpManager uses ICMP, SNMP, and TCP to check the availability of your devices and maintains an availability timeline for each device. You can customize the monitoring interval for uptime checks and set up alarms for non-availability. Moreover, you can also track the response time and packet loss to ensure quality as well as availability.

![Availability statistics and memory utilization in OpManager](https://cdn.manageengine.com/network-monitoring/sem/images/server-monitoring/server-monitoring-widgets1.svg)

### Get a glimpse of your server performance by tracking KPIs

**Keep track of server performance with dedicated, vendor-specific performance monitors** for metrics like CPU, memory, traffic, CPU temperature, and disk space. OpManager lets you set up three-level alarm thresholds for monitored metrics and generate context-rich alarms and graphs based on performance data.

![Performance monitoring in OpManager](https://cdn.manageengine.com/network-monitoring/sem/images/server-monitoring/server-monitoring-widgets3.svg)

### Manage critical services and processes effortlessly

**Get instant alarms for [process](https://www.manageengine.com/network-monitoring/process-monitoring.html?utm_source=seo&utm_content=server) and service outages.** You can use OpManager to fetch all the active processes and services running in your servers and set up regular monitoring checks for them. When any process or service is not running, you get an alarm specifying the outage, the process or service, and the device. You can also configure OpManager to eliminate and restart services or processes remotely.

![Monitoring the response time and KPIs for services and processes in OpManager](https://cdn.manageengine.com/network-monitoring/sem/images/server-monitoring/server-monitoring-widgets2.svg)

### Visualize server arrays and datacenters

**Keep an eye on server racks to enhance your monitoring capabilities.** You can map server racks with rack builders, map out datacenter floors, virtual hosts and VMs, layer 2 topologies, and more with OpManager's server mapping features. You can also leverage dashboards and graphs to analyze server performance with a quick glance. Dashboards and graphs colour-code and time-stamp server information to help you obtain quick, executable insights.

![Server rack views and datacenter visualization in OpManager](https://www.manageengine.com/network-monitoring/images/server-monitoring/server-monitoring-rack-view.avif)

### Leverage automated incident response to resolve server issues

**Prevent prolonged downtime and outages with OpManager's automated workflows.** You can create sequential, outcome-based, workflow actions. OpManager offers over 70 actions in a code-free, drag-and-drop workflow builder. All you have to do is arrange each action based on the outcome of the previous action to create an intuitive remediation workflow. When it comes to servers, you can use workflows to execute actions like restarting services, spinning up VMs, running scripts, and so on.

![Automated workflow response in OpManager](https://cdn.manageengine.com/network-monitoring/sem/images/server-monitoring/server-monitoring-widgets4.svg)

### Keep constant watch over your servers by monitoring logs

**[Monitor server logs](https://www.manageengine.com/network-monitoring/how-to-check-server-logs.html?server-monitoring) like Windows event logs and syslogs** to stay aware of your servers at all times. You can set up alarms for specific Windows event logs and create rules to alert you for sys logs that fall under the rules. You can also monitor application logs with preconfigured rules for Exchange, IIS, MS SQL, and ISA servers.

![Event log monitoring in OpManager](https://cdn.manageengine.com/network-monitoring/sem/images/server-monitoring/server-monitoring-widgets5.svg)

## What makes server monitoring with OpManager unique?

### ML-powered server management

OpManager features a powerful ML engine that uses advanced algorithms to predict the performance of the monitored systems. With adaptive thresholds, you can use OpManager to calculate, set, and adjust alarms thresholds for all monitored metrics. This improves anomaly detection and eliminates alarm flooding and false positives.

### Multi-vendor monitoring support

Monitor and manage your multi-vendor environment, including physical and virtual servers, effortlessly from a unified console with OpManager. Seamlessly manage physical servers from leading vendors like Dell, HP, IBM, Huawei, Oracle, and Lenovo as well as virtual servers from platforms such as VMware, Hyper-V, Nutanix, Xen, and Proxmox.

### All-in-one server monitoring (MSSQL, Exchange, DC, IPMI)

Whether your servers are used for running database operations, application hosting, mail, or domain management, OpManager can monitor them. You can monitor domain controllers, use IPMI to manage server hardware, track MS SQL transactions, [monitor Exchange](https://www.manageengine.com/network-monitoring/exchange-monitoring.html?utm_source=seo&utm_content=server) servers, and monitor different servers with dedicated metrics.

## What are the key server performance metrics to look for?

- **CPU or Memory Utilization:** Percentage of CPU or RAM in use. You can set alarm thresholds for higher values to prevent bottlenecks, poor performance, or resource scarcity.
- **Active Transactions:** Number of ongoing server operations. Helps identify load, potential deadlocks, or slow processes impacting performance.
- **Network Bandwidth Usage:** Amount of data transferred over the network. High usage indicates heavy traffic, saturation, or data transfer operations.
- **Data Files Size:** Cumulative size of application/database data files. You can track growth to perform capacity planning.
- **Packet Loss Rate:** Percentage of data packets failing to reach destination. High rates indicate network issues.
- **Active Database Connections:** Number of open database connections. High counts indicate application demand, connection pooling issues, or resource limits.
- **Interface Error Rate:** Percentage of errors on network interfaces. High rates suggest physical network problems, impacting server connectivity.
- **Logins per Second:** Rate of server authentications. Spikes can indicate high demand or potential security threats.
- **IO Read and Write Rate:** Speed of disk read/write operations. High rates suggest disk-bound applications or I/O bottlenecks.
- **Transactions per Second:** Number of completed server operations per second. Indicates server throughput and overall performance capacity.
- **Total Active Locks:** Number of held locks on server resources. High counts indicate contention, poor concurrency, or potential deadlocks.
- **Log Files Used Percentage:** Disk space consumed by log files. You can set up monitors to ensure log rotation, archiving, and to prevent disk space exhaustion.
- **Average Disk Latency:** Average time for disk I/O operations. High latency indicates slow disk performance, impacting application responsiveness.
- **Average Lock Wait Time:** Average time processes wait for locks. High wait times indicate resource contention, leading to performance degradation.
- **Partition Details of the Device:** Information about server disk partitions. Helps with capacity planning, identifying imbalances, and space management.
- **Number of Deadlocks per Second:** Frequency of deadlocks. High rates severely impact application performance and require immediate resolution.
- **Data Space of DB:** Total disk space consumed by the database. Monitor growth to perform capacity planning and efficient database management.
- **Instance Count:** Number of running application/service instances. Helps ensure scalability, resource allocation, and service availability.
- **Cache Hit Ratio:** Percentage of requests served from cache. High ratio indicates efficient caching, faster responses, and reduced load.
- **Thread Count:** Total number of active threads/processes. High counts can indicate resource contention, inefficient code, or memory exhaustion.

## How server monitoring works with OpManager

| Supported flavours for Windows servers | Supported flavours for Linux servers | Server monitoring protocols |
|---|---|---|
| Windows NT, 2000 | Red hat enterprise | SNMP v1 and v2c |
| Windows 2003 | SUSE | SNMP v3 |
| Windows 2008, 2008 R2 | Ubuntu | CLI |
| Windows 2012, 2012 R2 | Fedora | WMI |
| Windows 2016 | Debian | PrismAPI |
| Windows 2019 | CentOS | APIs |

## FAQs on server monitoring

### What are the server monitoring requirements in OpManager?

OpManager doesn't have any specific requirements for server monitoring. Servers are included in OpManager's regular device-based license. You can monitor unlimited metrics for each server in your license, including: uptime, performance, hardware, and the connected interfaces.

To discover, identify, classify, and monitor servers, OpManager uses:

- [SNMP](https://www.manageengine.com/network-monitoring/snmp-monitoring.html?utm_source=seo&utm_content=server)
- [WMI](https://www.manageengine.com/network-monitoring/wmi-monitoring.html?utm_source=seo&utm_content=server)
- CLI
- PrismAPI

You have to provide credentials for the respective servers for OpManager to reach and collect data from the servers. Alternatively, you can install a lightweight monitoring agent.

### What is the recommended monitoring interval for server monitoring?

OpManager lets you set up monitoring intervals ranging from 30 seconds to one hour or more for your devices. Depending upon the criticality of the server device, you can choose the right monitoring interval.

For instance, if your server is hosting a user-facing service that requires high availability and performance, you can monitor the server at 30 second intervals. This gives you a better chance to detect issues faster.

You also have to keep in mind that smaller monitoring intervals can increase the load on OpManager's monitoring server. If you add more than the recommended amount of low intervals, performance can be impacted.

OpManager also has an option to track specific metrics at 1-second intervals. These real-time "heart beat" monitors can be set up in dashboards to track mission-critical resources. For more information on how to monitor your servers, checkout this comprehensive [server monitoring checklist](https://www.manageengine.com/network-monitoring/blog/server-monitoring-checklist.html?server-monitoring-landing-page).

### How to set up alarms for server performance metrics in OpManager?

After discovering your devices, OpManager classifies them and creates a list of dedicated performance metrics based on the type of the server and the vendor. You enable monitoring for these metrics in bulk from a [device template](https://www.manageengine.com/network-monitoring/device-discovery/?utm_source=seo&utm_content=server) list, or individually, from the device's dedicated "snapshot" page. Once enabled, you can set up alarm thresholds. Based on the type of metric, you can set up various threshold conditions (Greater than, lesser than, equal to, etc.)

There are three levels of alarm severities for which you can set thresholds:

- Attention
- Trouble
- Critical
- You also have a rearm value that can be used to reset an alarm as "clear".

### How does ML-powered adaptive thresholds calculate thresholds for all metrics accurately?

OpManager uses machine learning to [automate the process of setting up alarms for various server performance metrics.](https://www.manageengine.com/network-monitoring/server-monitoring-automation.html?server-monitoring) Adaptive thresholds use algorithms to predict the "normal" or "expected" value for each metric. This is then set as thresholds for each metric and updated every hour automatically.

Adaptive thresholds needs up to 14 days of prior performance data to set alarms accurately, and as data collection continues, the accuracy increases. Given enough training data and time, adaptive thresholds can reach an accuracy of up to 90 - 95%.

### Other than uptime and performance, what can OpManager monitor for servers?

When it comes to servers OpManager can also monitor things like:

- **[Files and folders:](https://www.manageengine.com/network-monitoring/file-folder-monitor.html?utm_source=seo&utm_content=server)** Metrics like folder/file size, age, existence, and the presence of specific strings within files.
- **[URLs:](https://www.manageengine.com/network-monitoring/url-monitoring.html?utm_source=seo&utm_content=server)** Track the availability and check for specific strings.
- You can also create custom scripts using Perl, VB script, Powershell script, Shell script, etc and monitor specific metrics from your server devices.

### How does server rack views and datacenter floor views help?

[Rack views and datacenter floor views](https://www.manageengine.com/network-monitoring/datacenter-visual-modeling.html?utm_source=seo&utm_content=server) are server visualization tools used to track the location, uptime, and health of datacenter servers.

A rack view is a simulation of a server rack within which you can drag and drop server devices. Once racks are built, you can add them to a datacenter floor view. You can also add other things like walls and walkways in the floor view.

Once built, you have a visualization of your datacenter where each server can be tracked with a number on the rack and an alpha-numeric grid on the floor. So, if you want to direct your technician to a specific server, you can just tell them the exact location from the datacenter floor view. These floor views can also be rendered in 2D or 3D.

### Is OpManager suited for home servers or datacenter server models?

OpManager is versatile a server monitoring tool offering extensive monitoring features.

**For home servers:** You can use OpManager's standard or free edition to track uptime and basic performance metrics like CPU and memory.

**For datacenter servers:** You can use OpManager's professional or enterprise edition to ensure 99.999 percent availability, fault resilience, and automated remediation. You get visibility into server infrastructure, hardware, hosted VMs, application performance, and service delivery architectures from the network end to the application end.

### Why is OpManager considered a top choice for server monitoring?

OpManager is preferred by IT admins around the world for its simplicity, extensive feature offering, and affordable pricing.

- OpManager monitors 300+ dedicated performance metrics for 130+ server vendors, including physical and virtual servers, domain controllers, exchange servers, MS SQL, and more.
- Monitoring and licensing is intuitive and has little to no learning curve.
- Most importantly, OpManager's server monitoring features are included with its network performance monitoring feature sets and don't require additional licensing. This makes server monitoring with OpManager extremely cost-effective compared to other tools.

## Discover more on server monitoring

### Featured

- [OS monitoring](https://www.manageengine.com/network-monitoring/operating-system-monitoring.html?serv-mon)
- [Server Health Monitor](https://www.manageengine.com/network-monitoring/server-health-monitor.html?serv-mon)
- [Server Virtualization](https://www.manageengine.com/network-monitoring/server-virtualization.html?serv-mon)

### Quick links

- [Blogs](https://blogs.manageengine.com?hardware)
- [E-books](https://www.manageengine.com/network-monitoring/ebooks.html?serv-mon)
- [Videos](https://www.manageengine.com/network-monitoring/videos.html?serv-mon)
- [Case studies](https://www.manageengine.com/network-monitoring/customer-recommends.html?serv-mon)
- [Awards and Recognitions](https://www.manageengine.com/network-monitoring/network-software-review.html?serv-mon)

### Additional resources

- [Acing server performance](https://blogs.manageengine.com/network/opmanager/2023/09/22/acing-server-performance-dont-overlook-these-crucial-11-monitoring-metrics.html?serv-mon)
- [Network performance metrics](https://www.manageengine.com/network-monitoring/tech-topics/network-monitoring-metrics.html?serv-mon)
- [Associating Event log rule](https://www.manageengine.com/network-monitoring/help/associate-eventlog-rules.html?serv-mon)

## Related Products

- [Network Monitoring](https://www.manageengine.com/network-monitoring/?relPrd)
- [Bandwidth Monitoring & Traffic Analysis](https://www.manageengine.com/products/netflow/?relPrd)
- [Network Configuration Management](https://www.manageengine.com/network-configuration-manager/?relPrd)
- [Switch Port & IP Address Management](https://www.manageengine.com/products/oputils/?relPrd)
- [Firewall Management](https://www.manageengine.com/products/firewall/?relPrd)
- [Network Monitoring Software for MSPs](https://www.manageengine.com/network-monitoring-msp/?relPrd)
- [IT Operations Management](https://www.manageengine.com/it-operations-management/)
- [Application Performance Monitoring](https://www.manageengine.com/products/applications_manager/?relPrd)