Open topic with navigation
An important role of Analytics is to alert you when key resources within your GIS infrastructure stop functioning. For key servers or services, you can configure alarm notifications to be sent directly to phones via SMS or email so that you can act quickly to restore systems that are down. You can create a list of alarm notification recipients and then assign specific people to be notified when different resource alarms are triggered.
Alarm notifications are optional and configurable but all alarms are always displayed in Analytics.
The Summary tab of the Reports application is arranged specifically so that you see the most recent critical alarms when you first open the application. Most of the data in the Status tab is also arranged by resource and level of urgency to help you troubleshoot and get to the root cause of any problems within your system.
Similarly, the navigation, links and interactivity of each part of Analytics is designed to make it easy to drill down deeper into your GIS system to quickly get to detailed information about an alarm and its source.
An alarm and an error in Analytics are not equivalent, but have a very specific meaning based on how the information is collected.
When you are troubleshooting an alarm, you are likely to respond with a sequence of actions similar to those outlined below. The example is not an exact procedure but gives some sense of the steps you can take to help you find the source of a problem.
If you have set up alarm recipients to receive alarm notifications by phone or email, users are likely to begin the process after they receive either a text message (SMS) or an email notification of an alarm.
An example of how to investigate an alarm:
Open Analytics and examine the Summary tab.
Notice that it shows an alarm for a feature service called Pipeline_ZM.
Click the red Request Failure Alarm card.
Analytics opens the resource that caused the alarm to trigger:
There is nothing wrong with the ArcGIS Server alarms, so scroll down the page to the Services panel.
It is clear that the Pipline_ZM map and feature services have problems.
Click the alarm row to see more information about the alarm.
The row expands to show three tabs of information about this service.
It shows that the request to the service timed out, which is what caused the alarm, and the error has persisted for 6 days.
Click the Service Info tab.
The Service Info panel has an external link to the service causing the problem.
In Geocortex Analytics you can do any of the following with Alarms:
Check Now or Queue: You can run a check on an alarm at any time. The check that runs is exactly the same as the scheduled check on the resource. Windows Performance Alarms that are located on a Hub server or agent server cannot be checked immediately but need to be queued for checking during the next interval (usually less than 15 minutes).
To pause an alarm:
To pause an alarm, click the menu icon either in the alarm row or on the card describing the alarm, and then click Pause.
Click the Pause slider to activate it.
In the Paused Until box, use the calendar and time selectors to set a time interval that you want the pause to stay in effect.
In the row, the alarm icon changes to show this alarm is paused .
To turn off a pause on an alarm:
To end a pause on an alarm, click the menu icon either in the alarm row or on the card describing the alarm, and then click Pause in the menu.
Click the Pause slider to turn the pause off.
To run a check on an alarm:
To check an alarm, click the menu icon either in the alarm row or on the card describing the alarm, and then click Check Now or Queue.
Analytics runs a check on the alarm or the check is queued to run during the next available collection period.