v.69 Release: Services overhaul, two way integrations, and much more

As promised, the v.69 release includes a lot of great changes.  The most notable is the complete Services overhaul, which brings much more visibility to the multi step flow as well as improved data visualizations.   That shouldn’t overshadow the fact that two way integrations are also included, supporting the return of a ticket number for ConnectWise, AutoTask, and Custom HTTP integrations.  With this, we can update exiting tickets in your ticketing system based on alert status.  But that’s not all, new permissions controls for Help links, widget improvements, the ability to forecast performance from your dashboard, and much more are also included.  Read on for details.

The New Services

This release brings you an updated Services layout as well as improved workflows and visibility.  With this release, the old view of the Services page will no longer be available.

There are a lot of changes, all captured within the updated Services documentation.  Some highlights include:

General Functionality:

• The ability to globally manage testing locations.  Highlighting your account name in the Services tree will display a manage button, which gives you the ability to turn on or off testing locations for the entire account.  You can also enable or disable monitoring and alerting, and add Properties.
• That’s right, this release also brings properties for Services.  Initially Services properties can be used in alerts and reports.  We will extend this to make them usable within a service check.
• Yes, it also brings subgroups.  They function exactly as device subgroups, inheriting properties from a parent group.
• The layout itself is also more inline with the Devices layout, bringing Graphs, Alerts, and SDTs into a single level tab view.

Ping Checks

• Ping checks retain the ability to set the number of packets sent, but have a new configuration for percent returned and timeout.

Single Step Service Checks

• Single step service checks have been merged with multistep.  See multistep changes for more details.

Multistep Service Checks

Single and multistep have been melded into Add | Webservice.  Some changes you’ll notice:

• You are required to set a base URL.  Instead of www.google.com/subpage/index, you will need a URL of www.google.com, with step 1 of /subpage/index.  To change the base URL, you can select Service | Manage, changing the URL will automatically be reflected in all steps.
• Alert settings includes a total time setting.  This previously existed per step, which made it impossible to have a 5 step service check that you wanted to see complete in 2 seconds.  All steps must complete within the time set.
• There is a default time out of two minutes.  If you set a total time setting of 1 minute and the check does not complete, we will continue trying for another minute.  This enables us to give you performance information beyond a status alert.
• The combination of the above two features has played into an overall improvement on how response time is reported, now encompassing the time for the entire service check to complete.
• Many of the configurations in the check itself are the same, with a more intuitive layout.

Once your first step is created, you will be brought to the Steps tab.

From the steps tab, you can:

• Add more steps.
• Test a single step, or all steps.  This will display the request information sent, response information received, and time for the step to complete.
• Clicking the toggle to Off will disable the step.  The drop down gives you the options to edit, clone, or delete the step.
• Steps can be dragged and dropped to reorder.

Graphs and Data Display

• A new overall service graph is available, displaying response time per location.
• Individual location graphs include both a Response Time graph (time for all steps to complete), and Initial connection performance (Read, Wait, and DNS time for the first step).

If you are currently using Services checks, please take the time to read the following:

• A data migration as part of this release.  This may result in up to a two hour gap in historical performance data for multistep checks.  Status data will not be effected.
• There is only one setting for service completion time with this release.  If you have a multi step service check with a time configured per step, the total time allowed will default to the sum of time per step.
• Single and multi step checks have been merged into web service checks.  This will result in any check with more than a base URL (www.site.com/subpage) having a base URL of www.site.com, and a step 1 with /subpage.
• The ping service checks have default additional alert configurations for percent of packets returned, and time to return them.  The defaults are 80% and 500ms.
• You will notice a difference in the response time.  As the latest release includes a default timeout for the entire check and a response time for the entire check, the overall response time reported has been optimized to account for all elements of all steps.  
• If you have a services check that has not been updated since v.55 and requires authentication, an issue has been identified with the migration.  This will be resolved for all customers on Thursday, November 18 or can be updated manually.

Two Way Integrations – ConnectWise, AutoTask, and Custom options

To better capture what Custom Alert Delivery Methods actually do, they have been renamed Integrations.  You can access them from Settings | Integrations.  Our other integrations, including ConnectWise and AutoTask, will be accessible from the same screen.  

All integrations now have the ability to retrieve a ticket ID, or unique identifier, from the external system.

This enables you to include the ticket ID anytime an alert is acknowledged, escalated, or cleared – effectively updating existing tickets based on the LogicMonitor alert status.

The alert data can be configured per change type:

Due to the changes in functionality, this will only no longer be available in the old UI view. More details about specific integration configuration are available here.  Other integrations, including PagerDuty and ServiceNow are in the works.  

More Role Based Access Control Options

From Settings | User Access | Users and Roles, there are new configurations for setting role permissions.

In Roles | Settings, there is an option for Can edit personal information.

This is only available if Manage User Access is not selected.  If not checked, the user will not see the Edit Profile option when clicking on their username.

If checked, the user will be able to edit name, password, and phone number.

There is also a new section for Help permissions.  You can set which options will display for users when they click the help dropdown in the main navigation bar.  If no options are selected, will not display.

You can also add a custom help link, if you’d like users to contact you directly with questions, or if there i
s an internal help site.  

If a user has multiple roles with different custom links, all link options will display.

Forecast from the Dashboard

Take advantage of the power of capacity forecasting directly from your dashboard.  From any graph widget, click the drop down.  Forecast will be one of the options:

New NOC Widget Options

Within the NOC configuration, you will find two new options for which alert severities will display – Acknowledged and In Scheduled Down Time.

If Acknowledged is checked, an item where the highest severity alert is acknowledged will appear in the NOC Widget:

If unchecked, the item will display as green.  Similarly for In Scheduled Down Time.  If unchecked, alerts in scheduled down time will be ignored in determining status for the NOC widget display.

Acknowledged and In Scheduled Down Time will be selected by default, as this was the initial behavior.

Pie Chart Options

An option has been added to pie charts to enable you to limit the number of slices in the pie.  If you limit the results, you can opt to display all remaining values as a sum with the label “Others.”  

If you choose to show all and have more than 25 items, only the top 25 will display.

Other Improvements:

• When editing widgets or configurations that allow for data rows to be saved independently, the row will be automatically saved if the main Save button is selected.
• The OpsNotes layout has been improved, displaying in a larger frame to account for in depth notes or many tags.  Selecting a row will enable you to make edits.
• A new token, ##DEVICEURL## will resolve to the Link to URL information configured on a device.  This token can be used within alerts and reports.
• Single Sign On has been moved under User Access in the Settings page.  This is a more appropriate location.  Please note that role based access will be dependent on the ability to view and manage the User Access section, not the former Integrations section.
• Ongoing effort to improve dashboard speed and widget display times.  Please let us know if you’re seeing speed issues, examples often help us with our optimizations.
• A link to Amazon’s Status Website has been added from the main AWS group | Info tab.  We’ve also included device counts from unmonitored regions in the totals table, to give you a full view of what is included in your group.
• Access log entries have been enhanced to include more details if a dashboard widget is moved between dashboards, or if a device is moved between groups.  The default time range has also been changed to 30 days.
• A new property is available, system.collector, to indicate a devices hosts a collector.  This will be used for datasources, and will be automatically applied with future collector versions.
• Top ten overview graphs will now include data from deleted instances.  A * will be appended to the instance name if it has been deleted.
• On the device dashboard, a link to raw data has been added for single instances.
• A detailed error has been added when a duplicate instance is added from Add Other Monitoring.  Previously the addition failed without a clear explanation.
• Added total count of alerts to the Alert Tab for groups.
• Improvements have been made to the Began column for the alert table, displaying all alert times in a 24 hour format.
• Search on Settings | Users and Roles will search all applicable fields, including name, email, and role.  The search applies to all users, and will refresh to page 1.
• The network traffic flow tables continue to get some minor cleanup.  In this version, you’ll notice that the IP and Description columns have been combined – e.g. mydevice (10.0.0.1) for Top Talkers, and Description and Port have been combined – e.g., https (443), for Top Applications. Only the port or IP will display if there is no description.  The time range filter has also been removed, as we’ve learned that the time range for the application is more commonly used.
• Better messaging has been added when trying to Add Other Monitoring | Add Processes.  Previously a generic error displayed without indication if the processes were loading slowly or if they were not available.
• We’ve made some improvements to how disabled alerts are communicated throughout the Device page.  If alerts are disabled for the entire group or device, it will display in the upper right for the datasource.  If datapoint alerts are disabled, all groups will be listed in the instance | Alert Tuning tab.
• ACH payments can now be added in Settings | Account Information.
• A new column, Instance Description, is available for the Alerts Table.  This can be found in the widgets, device alerts, and alerts page.
• Additional details have been added to the AWS policy errors.  If a AWS service is selected that is not correctly supported by the AWS user credentials, specific details on required permissions are displayed.
• The ability to copy has been added to all widget datapoint configurations.
• Long report names will wrap to fit within the Report Summary box on the reports page.
• Device search results will be returned by best match and then alphabetical order.

Bug Fixes:

• Alerts were not displaying for instances with special characters in the name, such as ‘/’ for drives.
• Searching for cluster will return all cluster alerts, it no longer requires ‘<‘ or ” to match.
• Widgets displaying data for instances with a one hour polling interval or greater could result in empty data due to how values were cached.  
• Addressed an issue with the task that evaluated which datasources should apply to a device, occasionally requiring the manual selection of “Force datasource rematch.”  
• Unique From numbers have been revalidated for each account.  This may result in LogicMonitor sms and voice alerts coming from a new phone number.
• Device descriptions were not displayed correctly within the google map widget.
• The total displayed on the Alerts Tab in Devices was not always refreshed when the table was updated.
• Graph autoscaling was not working correctly if OpsNotes were visible.
• If a dashboard name was edited, the prior name was not always over written.  This has been fixed.
• If your NOC widget is sorted by severity, the items within each severity will be displayed in alphabetical order.  
• When configuring widget datapoints, we started the Datapoint summary started at devices.  Although you selected Group 1 | All devices, the summary would say All Devices > All Datasources > …  Group is now included in the summary.
• Using a non alpha numeric character in the Pie Widget configuration returned an unknown error.  This has been clarified to indicate special characters are not allowed.
• Alert icons were not instantly refreshing if an alert was acknowledged from the top level of the device tree.
• Fixed a few role based access display bugs, including displaying the NOC configuration for View Only roles, displaying the OpsNotes button for non admins, and displaying a permission denied error when a user with view rights tried to open alert rules.

Other items of note:  

• As of this release, support for sFlow version 2 has been officially deprecated. 
• If your account name is changed and you have a existing dashboard with the new account name, a ‘_’ will be appended to the end of the dashboard name to create a unique value.
• We’ve added support for native powershell scripting in datasource configuration.  This has a collector dependency, and will be detailed with the next EA collector release.
• A change has been made to the GetAlerts RPC that includes a limit on the number of alerts that can be retrieved, to avoid performance issues.  More details available here.