Monitors included into Server health checks – Windows server template
Server health checks – Windows server application template provides monitors for core hardware and software resources, vital for normal server operation. The monitors marked enabled by default (listed in bold below) are those we strongly recommend to leave enabled. The rest of them might be more or less important, depending on actual server usage. You are advised to enable them according to services and resources supported by the server to be monitored. More about templates.
The below monitors are part of Windows Server template (Server health checks category):
CPU Usage (total active time) (enabled by default). CPU is a scarce resource; it’s essential to watch its usage and notice any overcommitment in time. It’s recommended to keep the monitor running.
Free disk space on disk C: – disk C: is usually the storage device used by most programs and system needs. Lack of available space may easily result in system malfunction or even failure. Although this monitor isn’t enabled by default, it’s commonly recommended to keep it running. Make sure you have at least a few gigabytes free on that drive, at any moment.
Free physical memory – RAM (physical memory) is most precious resource; its lack will result in frequent swapping in and out activity, thus in significant performance decrease of the entire system. Make sure you have at least 15-20% of RAM always free and set up alerts to inform you if not.
Hotfixes installed shows the number of OS/components updates having been installed since the moment of its last restart. OS offers automated installation/restart of hotfixes, since they can mean important security and/or efficiency related changes.
PING (enabled by default). Simplest network administration tool, ICMP Ping is still reliable means to detect connectivity and general availability problems. It’s often used as dependency monitor, used to prevent other monitors from sending multiple Down alerts, if PING goes Down.
Page file usage – page file, also known as swap file, is used to temporarily exchange (swap) contents of RAM between processes – for example, of process is requiring RAM not currently available in requested amount. In most cases, intensive usage of page file indicates that it’s insufficient RAM installed, or there are applications that could perhaps be stopped to release RAM they used.
Reboot required (updates installed) – most important hotfixes often require system reboot, to be applied properly. This monitor can help to indicate that – if reboot is required, it’s recommended to perform it as soon a possible.
Process count shows the total number of running processes.
Traffic speed total, kbit/s – traffic accounting is an important part of network administration. Make sure there are no runaway processes that consume too much traffic. Run the monitor for a while, to determine which average traffic consumption can be considered normal – to set up state conditions properly.
Traffic volume total – another aspect of traffic accounting, namely the overall traffic volume monitoring. Sudden spikes in traffic consumption may indicate there’s unnecessary activity happening on the server.
Uptime – general metrics of device availability in the course of time. High uptime values mean the device is well connected, and its owners can handle the server problems. Note: keeping very long uptime is generally a bad idea, since devices might become too vulnerable.
Windows Service: Lanman – useful if your computer runs Server service (i.e., provides Windows Server resources). If you don’t use that service, this monitor is not necessary.
Windows Service: Netman – essential part of all the software using networks. If the monitor is Down, it should use a reliable alternate alerting means, such as GSM modem. Otherwise, if cut off network, monitor will be unavailable to inform about important updates.
RDP – typical means to access Windows installation from outside. Usually is granted for servers. Please make sure RDP is enabled, especially if you’re installed computer remotely – error in its configuration may render the computer inaccessible.
RPC Endpoint Mapper – allows software to perform remote calls on the computer. Note that this is required, if calls using, say, WMI, all expect the mentioned service. If you have firewalled access to it, make sure that are enough exceptions defined.
Server health checks – Windows server use cases
- for file servers, disk capacity and network throughput (traffic speed) are crucial; make sure both those monitors are set up and trigger alarm, if free disk space becomes too low, or bandwidth speed rises far above the expected average
- for domain controllers, CPU usage, memory usage and disk usage are all vital monitors; any outage, or state close to outage, should be promptly warned about: domain controllers are often authentication sources; when Down, they can cause many other local network services to halt
- if free memory and page file usage monitors are at least in Warning state, it can indicate there are memory leaks and/or system problems – time to check what is using that much memory; these monitors are the must for every development server
Server health checks – Windows server tips
- although uptime shows the stability of the server, it also can indicate it is long missing important updates; make sure both Uptime and Hotfixes installed are enabled and working, to get warned in time that Windows requires maintenance
- PING monitor should be the one other monitors depend upon; make sure it warns administrators, when down, often enough, but let other monitors be stopped by dependency, in order not to send too many alerts
- Process count is a must for every high-availability system: make sure the processes are running during time they are expected to be up, and try both alerting network administrator and re-launching the stopped application, if important processes are all stopped; note you might need to add several monitors of this kind, to cover all the important processes
- for every remote Windows server RDP and Windows Service: Netman monitors are a must; as an action to its Down state, try re-starting Remote Desktop Windows service, to prevent losing connectivity with the server
- if WMI or any other remote scripting is in use, make sure RPC Endpoint Mappers monitor is enabled and in OK state; that also relates to any other Windows system that is probed or managed remotely
IPHost Network Monitor provides application templates (or just “templates” later in document), to create multiple relevant monitors in only a few clicks. Templates facilitate adding typical monitors sets; this can be particularly useful in case of big networks, when creating same-type monitors for many same-type devices is a common task. Application templates are sets of monitors that can be added, using specific predefined parameters, for a given host at once. The said set, added for given host, is displayed as a separate node in tree view pane, and is named application.
There are predefined templates; user can as well generate templates of their own – either out of existing monitors, or by cloning a predefined template. User-added template definitions are saved in XML files and can thus be conveniently augmented or applied to specific needs.