The Event list uses grid-like UI to show you an ordered list of all the events within the current timeframe and view. Each row in the grid describes a discrete event and provides deep contextual information as to its type, location in the code and impact in terms of volume and spread across your application.
For each event the grid shows the following values:
- Type - can be a Caught Exception, Uncaught Exception, Log Error, Log Warning, HTTP Error, Custom API Event.
- Error - the name of the error. In case of an exception this would be the exception class name (e.g. AmazonException, NullPointerException, etc.). In case of a log error or warning this will be a sample message for that event (e.g. "Could not complete operation for user... ").
- Error Location - the application class and method in which the event occurs. Even if the actual error is generated by a 3rd party or core JDK frameworks, OverOps will show the last location in the application code calling into the error (vs. 3rd party code), to help you focus on the code that's being impacted within your application. It is also possible to filter on packages.
- Entry Point - the application class and method name at the top of the call stack leading to the event. This enables you to see the business transaction / application flow which this error impacts.
- Times - the number of times this event has occurred. This value is key to informing you of the severity of the error to answer questions such as "has it happened once or a million times over the last hour?" Notice that the Times column adjusts itself to the current timeframe, meaning it will dynamically reflect the number of times this error has occurred in the current timeframe (e.g. last hour, last day, 10:30 - 11:00pm, etc.).
- Error Location Calls - the number of times the application method containing the error has been called. For example, an application method foo may be calling into framework util which is called into by many locations within the application. OverOps measures the number calls specifically into foo vs. util, which may contain many calls that are irrelevant to this error. This is especially helpful within the context of the Error Rate column below.
- Error Rate - the percentage of time when this errors occurs out of calls into the error location. This value is calculated as Times / Error Location Calls. This value provides powerful context as to how many times this occurs compared to calls into the actual code to enable you to answer questions such as - "is this error happening 2% or 50% of the time?".
- First Seen - the first time in which this error was first detected within your environment. This column is helpful in answering questions such as - "is this error is new and was just introduced as part of the new release, or has been there for a long time now?".
- Last Seen - the last time this error was detected within your environment. This value is key in letting you know whether this error is still impacting your application(s).
- Applications - lists the applications (e.g. "Producer-Service", "Consumer-Service", Web-frontend", etc.) in which this error was detected. Click here to learn how to assign Application names to your JVMs.
- Servers - lists the servers in which this error was detected. Click here to learn how to assign names to your monitored servers.
You can hover the Applications and Servers column to see which Servers or Applications have reported this error:
The events grid can be searched and sorted through to help you focus on the errors that are most relevant to you. This serves a strong purpose as it enables you to answer powerful question related to each column. Here are some common examples -
- Sort by First Seen - shows you the newest errors introduced into your application(s) within the last day, week... This enables you to know exactly when new errors have been introduced into your environment.
- Sort by Times - Shows you the errors happening most within the current timeframe.
- Sort by Last seen - when combined with viewing the Last day/hour timeframe, sorting by Last seen enables you to see the errors happening within your application(s) in real-time.
- Filter Type to Uncaught Exceptions - this filter enables you to see which exceptions and not handled by your code, which may lead to undetected failures
- Filter by Entry Point or Error Location - this filter enables you to view events that opriginate are are called into from specific locations in your code such as key t
- Filter by Server / Application - this filter enables you to see the impact and the levels of errors impacting a specific set of monitored servers or JVMs which you've tagged with applications names. Click here to learn more about tagging servers and JVMs.
* You can use the and at the top of each column to sort and filter events.
When you set filters within the gird they will be automatically reflected in the Filter bar which will appear underneath the grid header.
You can also save your filters as Views which can be reused and shared with teammates to help you directly focus on the errors and applications most critical to you. Click here to learn more about Views.
You can pick which columns to show using the Column Picker on the top right of the grid. Your choice will be saved as part of your view settings.
You can use the Search box to search for any any error type, name, location and entry point within your application. The search box will auto-complete phrases for to further make searching easier.