The following is a list of the kinds of alarms that can be configured in Analytics for various resources.
ArcGIS Services Alarm
When an ArcGIS Server is monitored by Analytics, by default, all the services on that ArcGIS Server are also monitored and will have alarms associated with each service.
If other resources in Analytics, for example a Portal or an Essentials instance, use a service on that ArcGIS Server, it too will be monitored and have alarms that you can configure.
You can also enable any service associated with a specific resource that is not automatically monitored. When you enable that service, alarms are added to the service and you can then configure those alarms to trigger under specified conditions. For example, you can enable a service associated with an Essentials Site, and then configure a Request Failure alarm to trigger when that service is not available.
If there is a service that is important to your infrastructure, but you do not own the service or it is not on one of your ArcGIS Servers, you can enable the service so Analytics can monitor it by finding it via the resource that is referencing that service:
ArcGIS Online instances
Portal for ArcGIS instances
Essentials Sites
Enabling a service for monitoring adds to the load of an ArcGIS Server. Analytics manages the load on your own monitored ArcGIS Servers by using Intelligent Probing but it is not able to intelligently monitor external services. Do not enable many services on the same ArcGIS Server if you do not own it.
To enable an ArcGIS Service alarm:
In Configuration, navigate to a specific resource under either Portals, ArcGIS Servers, Essentials or WebOffice.
Select the Alarms tab and scroll down to the ArcGIS Services section.
In the ArcGIS Service drop-down list, select the service you want to enable.
If you have many services, there may be a delay while Analytics finds and populates the list.
Ensure that the service is enabled.
If you want to set a default time interval between data collections for all the alarms, select a number, and then select either Minutes or Hours.
Click Save.
Once the Service has been saved, scroll down to see the Request Failure and Response Time alarms.
If Analytics is configured to monitor the parent ArcGIS Server, you will also see CPU and Memory usage alarms. Feature services use the same processes as their parent map service, so feature services will not have cpu or memory alarms.
The Available Storage alarm allows you to monitor when the storage levels of Portal for ArcGIS instances go below an acceptable level. You set a minimum value as a percentage below which available storage must not go. The alarm triggers if the storage on that resource goes below that percentage. The default is set to 10%.
Configuration| Portals | Portal for ArcGIS instance | Alarms tab.
Available User Accounts Alarm
The Available User Accounts alarm allows you to monitor when the number of available user accounts on ArcGIS Online and Portal for ArcGIS instances get below an acceptable level. You set a value for available user accounts on ArcGIS Online or Portal for ArcGIS. If the available user accounts go below this value, the alarm triggers.
The default is set to 10 users.
Resources this alarm can be set on:
Configuration| Portals | Alarms tab (ArcGIS Online and Portal for ArcGIS).
Configuration| Portals | Portal for ArcGIS instance | Alarms tab.
Content Match Alarm
The Content Match alarm allows you to monitor the content of websites and trigger an alarm if specific content is present or not present. Analytics sends the website a request and if the content returned does not meet the set criteria, then the alarm triggers.
Alarm Settings:
Click to open the Content Match Alarm and change the following:
Type: Select the condition of the content you want to trigger the alarm. Once you have selected this option, the subsequent options in the dialog change to match your selection. The options include:
Content is empty: The alarm triggers if no content is found on the website, even though the request was correct and may have been successful.
Contains: The alarm triggers if specific content is found on the content returned by the website.
Value to Match: The specific text that will trigger the alarm if it is found in the content.
Occurrences: Set the number of times the Contains criteria must be met before the alarm triggers.
Does not contain: The alarm triggers if specific content is not found in the response returned by the website.
Value to Match: The specific text that will trigger the alarm if it is not found in the content.
Occurrences: Set the number of times the Does Not Contain criteria can be met before the alarm triggers.
Resources this alarm can be set on:
Configuration | Websites | Alarms tab (affects all the websites Analytics monitors)
The Database Unavailable alarm allows you to monitor key databases so that if they go down, you can take quick action to restore them. Like other alarms, you can also add a script to run automatically when this alarm triggers, for example the script could reboot the server.
The Maximum CPU Usage alarm allows you to monitor when any resource uses too much of the CPU. You set a maximum usage value beyond which a resource must not go. If the resource uses more than the configured percentage, the alarm triggers.
Alarm Settings:
Click to open the Maximum CPU Usage Alarm and change the following:
Percent: Select the percentage of the CPU usage this resource cannot go beyond. The default is 80%.
Occurrences: Set the number of times the Percent criteria can be met before the alarm triggers.
Resources this alarm can be set on:
The Maximum CPU Usage Alarm can be set on ArcGIS Server (including individual services), Portal for ArcGIS, and Essentials. For ArcGIS Server and Portal for ArcGIS, an Analytics agent must be installed on the same machine or a remotely monitored host server must be associated with the resource. The CPU usage of an individual ArcGIS service can also be monitored if it is part of an ArcGIS Server that meets this criteria. Feature services are an exception, as they use the same process as the associated map service and so the CPU usage of a feature service cannot be monitored.
Maximum Memory Usage Alarm
The Maximum Memory Usage alarm allows you to monitor when any resource uses too much memory (RAM). You set a maximum usage value beyond which a resource must not go. If the resource uses more than the configured percentage, the alarm triggers.
Alarm Settings:
Click to open the Maximum Memory Usage alarm and change the following:
Percent: Select the percentage of the Memory usage this resource cannot go beyond. The default is 80%.
Occurrences: Set the number of times the Percent criteria can be met before the alarm triggers.
Resources this alarm can be set on:
The Maximum Memory Alarm can be set on ArcGIS Server (including individual services), Portal for ArcGIS, and Essentials. For ArcGIS Server and Portal for ArcGIS, an Analytics agent must be installed on the same machine or a remotely monitored host server must be associated with the resource. The memory usage of an individual ArcGIS service can also be monitored if it is part of an ArcGIS Server that meets this criteria. Feature services are an exception, as they use the same process as the associated map service and so the memory usage of a feature service cannot be monitored.
Query Failure Alarm
The Query Failure alarm triggers when a database query fails for any reason.
Configuration | Databases | [Database Name] | Queries tab (expand a query to view its alarms).
Remaining Credits Alarm
The Remaining Credits alarm makes it possible to monitor usage of your ArcGIS Online credits and notify key users when they get below an acceptable level. You set a value for remaining credits in the alarm. If the credits go below this value, the alarm triggers. The default is 2000 credits.
The Request Failure alarm allows you to monitor when any web-based resource does not respond to an HTTP request to monitor its status. When the resource fails to respond to the request, the alarm triggers.
Resources this alarm can be set on:
All web-based resources.
Response Time Alarm
The Response Time alarm allows you to monitor when a web-based resource takes too long to respond to an HTTP request. You set an acceptable response time for each resource. If the resource takes longer than that number of seconds to respond, the alarm triggers. The default is 30 seconds.
Alarm Settings:
Click to open a Response Time alarm and change the following:
Seconds: Set the number of seconds within which this resource must respond.
Occurrences: Set the number of times the seconds criteria can be met before the alarm triggers.
Resources this alarm can be set on:
All web-based resources.
Server Unavailable Alarm
The Server Unavailable alarm allows you to monitor key servers so that if they go down, you can take quick action to restore them. Like other alarms, you can also add a script to run automatically when this alarm triggers, for example the script could reboot the server.
On the Hub server, there is no Server Unavailable alarm because if the Hub server is down, information about that server cannot be collected or displayed.
Analytics collects Windows Performance data about the servers it monitors, for example, processor activity, physical and virtual memory, and many others. Much of this information is based on the Windows Performance Monitor, which you can access by clicking the Start menu on the desktop and typing Performance Monitor.
In Analytics, you can set an alarm to trigger on the performance of a Windows Counter so that if one of the counters is particularly important to the health of your system, you will be notified when there is a problem. Not every performance counter can be set up with an alarm. See the table below for the counters that can be added to Analytics.
To find information on Windows counters, open Performance Monitor on your desktop, then in the tree on the left, expand Monitoring Tools and click Performance Monitor. In the table at the bottom of the window, right-click the Counter column and select Add Counters. When the Add Counters window opens, you can select Show Description and then click on any counter in the list to see a description of what information the counter collects.
Analytics Name for the Category of Counter
Performance Monitor Name
Description (adapted from Performance Monitor)
Process
Process
Monitors running applications and system processes.
Memory
Memory
Counters that describe the behavior of physical and virtual memory on the computer. Physical memory is the amount of random access memory on the computer. Virtual memory consists of the space in physical memory and on disk. Many of the memory counters monitor paging, which is the movement of pages of code and data between disk and physical memory. Excessive paging, a symptom of a memory shortage, can cause delays which interfere with all system processes.
Physical Disk
Physical Disk
Counters that monitor hard or fixed disk drive on a computer. Disks are used to store file, program, and paging data and are read to retrieve these items, and written to record changes to them.
The physical disk counters are a sum of the values of the logical disks (or partitions) into which they are divided.
Network Adapter
Network Adapter
Consists of counters that measure the rates at which bytes and packets are sent and received over a physical or virtual network connection. It includes counters that monitor connection errors.
Logical Disk
LogicalDisk
Counters that monitor the logical partitions of hard or fixed disk drives. Performance Monitor identifies logical disks by their a drive letter, such as C.
Processor
Processor
Monitors all the activities of all processors including computations, and operations on other hardware like printers. It also runs the threads of processes. Each processor is a different instance.
Server
Server
Counters that measure communication between the local computer and the network.
System
System
Counters that apply to more than one instance of a component processors on the computer.
Network Adapter
Network Adapter
Counters that measure the rates at which bytes and packets are sent and received over a physical or virtual network connection. It includes counters that monitor connection errors.
IIS App Pools
W3VC_W3WP
Exposes the HTTP request processing related counters for the worker process.
IIS Sites
Web Service
Counters specific to the World Wide Web Publishing Service.
Alarm Settings:
On the Alarms tab of a specific server, in the Windows Performance Counter Alarms section, click Add Alarm.
Display Name: Enter the name of this alarm as it will appear in Analytics.
Category: Select a counter category (see the table above for options). The options in the Add Alarm window change to match the Category you selected.
Counter: Select the counter you want to set the alarm on.
Instance: The instance of the object you want to set an alarm on, for example, the IIS Site, the physical disk, network adapter, etc.
Minimum / Maximum: The maximum and minimum values for the performance of the counter. If the value collected by Analytics is below the minimum or above the maximum, then the alarm will trigger.
Occurrences: Set how many times the criteria should be met before the alarm triggers.
Script to run: Select a batch script to run (where applicable).
Email Recipients : Select an Alarm Recipient (optional) to receive an email notification when the alarm triggers.
SMS Recipients: Select an Alarm Recipient (optional) to receive an SMS notification when the alarm triggers.
to open the Content Match Alarm and change the following:
