Release Date: May 2019

Dashboards and User Interface

Firmware column in infrastructure pages

Firmware version on a device is now available as a column in the device listing page.

Firmware column in infrastructure pages

Firmware column in infrastructure pages

URL for resources

The details page for each resource now has it own unique URL that is based on its resource ID. You can directly access the details page of a resource via this URL.

URL for resources

URL for resources

Provided more categorized view of alerts in device overview page (in New UI)

Provided more categorized view of alerts in device overview page

Provided more categorized view of alerts in device overview page

Provided more categorized view of alerts in device overview page

Provided more categorized view of alerts in device overview page

The user can view alert categorization on the Device Overview page.

Critical, warning, and total. Open, acknowledged and Ticketed details can also be seen.

Changed the design (style) of the popup across the portal to match the new Infrastructure style

Old UI:

Changed the design (style) of the popup

Changed the design (style) of the popup

New UI:

Create New Maintenance Period popup has a new UI

Create New Maintenance Period popup has a new UI

The Create New Maintenance Period popup has a new UI.

Search a service map with capital letters is not working in the service maps listing page

Search a service map with capital letters is not working

Search a service map with capital letters is not working

Issue is fixed. The service map is displayed according to the search text with capital letters.

Gateway Profile assignment option is missing for cloud DB resources in New UI

Gateway Profile assignment option is missing for the cloud DB resources

Gateway Profile assignment option is missing for the cloud DB resources

Issue is fixed. The option can be seen for the Cloud DB resources like AWS RDS in the New UI.

Advance Search in Schedule maintenance Tab of New UI is not working

Advance Search is not working, for all the statuses: Active, Pending, Completed and Suspended.

Even if there are schedule maintenance created, the list is empty.

Advance Search in Schedule maintenance Tab of New UI is not working

Advance Search in Schedule maintenance Tab of New UI is not working

Issue is fixed.

Advance Search is working perfectly showing all the records, for all the statuses.

504 Gateway Time-out error is displayed

When trying to get the graphs for the interface on the devices (PDC and SDC), 504 error is displayed.

504 Gateway Time-out error is displayed

504 Gateway Time-out error is displayed

It has multiple cassandra db calls retrieving data logs against the interface metrics in a single request. It takes more time to process those multiple db requests if it has more data, so it takes more response time, and hence 504 gateway timeout error.

Loading the monitor values after the page is loaded. Issue is fixed.

There is no issue with graphs and interfaces for both the PDC and SDC devices.

Service Groups are not responding in Service Maps, in the New UI, in different browsers

When adding new groups to the Network service group in the new UI, all of a sudden the group no longer displayed anything.

504 Gateway Time-out error is displayed

504 Gateway Time-out error is displayed

Filtered out unnecessary mappings created through service groups update calls. Issue is fixed.

Able to view the network service maps. This is working properly in all the browsers.

Service maps side panel is not displaying service impacting resources and Service impacting alerts accordingly

Side panel of Service Maps is showing all the resources. It should show only the impacted resources that are causing service group to be Down, under the Summary tab or when a parent node is selected.

Service maps side panel is not displaying service impacting resources and Service impacting alerts

Service maps side panel is not displaying service impacting resources and Service impacting alerts

Issue is fixed.

The service impacting resources are only showing based on the status of the service node or device node. The impacting resources will only be shown if the node is down and/or degraded.

When the user clicks on the impacted alerts, a small window opens that shows the alert description along with tickets and availability alerts.

Service and Topology Maps

Indicator for shared services

Services that are members of multiple are now indicated by an icon, making it each to identify such common services.

Indicator for shared services

Indicator for shared services

Alert Management

Component name in alert details

Alerts from components, such as disks and network interfaces, now show the component name.

Component name in alert details

Component name in alert details

Improved inference context with incident

For incidents created from an inference, you can now see resources associated with each alert that was correlated with the inference. The Resources and Activity Log tabs within the incident show associated resources. This helps investigate the incident faster.

Improved inference context with incident

Improved inference context with incident

Improved inference context with incident

Improved inference context with incident

Additional condition options in escalation policy

You can now define alert escalation conditions based on resource attributes, in addition to alert attributes. The corresponding API has been updated to reflect these new options.

Additional condition options in escalation policy

Additional condition options in escalation policy

Heal and Suppress Inference when all of its alerts are de-correlated

Heal and Suppress Inference when all of its alerts are de-correlated

Heal and Suppress Inference when all of its alerts are de-correlated

If the inference alert was not suppressed which has no correlated alerts, it would remain in Open state.

The alert is healed; however, the incident is still in the New state

The alert is healed; however, the incident is still in the New state

The alert is healed; however, the incident is still in the New state

Issue is fixed.

Incident is getting created based upon the Escalation policy when alert got created and moving to resolved state once alert got healed.

Cloud and Cloud Native Environments

Microsoft Azure Application Insights metrics

OpsRamp can now collect metrics from Azure Application Insights. With this capability, you can now use OpsRamp to monitor application metrics from across your Azure accounts.

Microsoft Azure Application Insights metrics

Microsoft Azure Application Insights metrics

Support for AWS Aurora Serverless

You can now discover and manage AWS Aurora Serverless.

Support for AWS Aurora Serverless

Support for AWS Aurora Serverless

Support provided to build out a node for SQL Server or AppEngine

Account ID is provided as an option to select.

Support provided to build out a node for SQL Server or AppEngine

Support provided to build out a node for SQL Server or AppEngine

Resource Attributes

New resource attribute – Alternate IP

Compute resources now have a new attribute called Alternate IP. This attribute can be used to associate a non-discoverable IP address, such as a NAT IP address, with a resource. Once the user associates an Alternate IP address with a resource, Ping monitors applied to the resource will automatically monitor the device using the Alternate IP address, if the resource is unreachable via its primary IP address.

New resource attribute – Alternate IP

New resource attribute – Alternate IP

New resource attribute – Alternate IP

New resource attribute – Alternate IP

Service Management

First Response Time in incident header

You can now see an incident’s First Response time in the header. This makes it easy to track incident response against the defined SLA.

First Response Time in incident header

First Response Time in incident header

When an email is sent in JP (Japanese), subject and description of the incident is garbled

As seen in the below image, the incident is created, but the subject and the description information (Japanese characters) is garbled.

Subject and description of the incident is garbled

Subject and description of the incident is garbled

Issue is fixed.

The E-mail Subject and description appear fine without any garbled characters.

Knowledge Base

Support provided for searching articles that are going to expire within a week, two weeks or a month

Support provided for searching articles

Support provided for searching articles

The user can search for articles that are expiring in a week, 2 weeks or a month from the date of login.

Reports

Added Asset First Managed Time and Asset Managed time fields in custom inventory report

Added Asset First Managed Time and Asset Managed time fields

Added Asset First Managed Time and Asset Managed time fields

Renamed Synthetic to Web Services in Standard Report and Report Composer

Renamed Synthetic to Web Services

Renamed Synthetic to Web Services

Renamed Synthetic Resource Type to Web Services in Reports creation page and in reports data.

Patches can be viewed in the report for selected time for all devices

User can view the patches in the report for selected time for all devices, including device type KVM when filter rule is “Install status equals Installed.”

Patches can be viewed in the report for selected time for all devices

Patches can be viewed in the report for selected time for all devices

Patch Management

Added Show Matching Members option, while defining Patch baseline, on Add Patch Baseline screen

Added Show Matching Members option while defining Patch baseline

Added Show Matching Members option while defining Patch baseline

The patches that match the filter criteria are displayed.

The user can view the matching members (results) before saving the baseline.

Monitoring Management

Added Application Type filter in Metrics and Monitors list

Added Application Type filter in Metrics and Monitors list

Added Application Type filter in Metrics and Monitors list

Added Application Type filter in Metrics and Monitors list

Added Application Type filter in Metrics and Monitors list

Added Application Type filter in Metrics and Monitors list

Added Application Type filter in Metrics and Monitors list

Application Type is added as a column in both Monitors and Metrics listing.

Integrations

Full unique id of the integration event is displayed when the user hovers over the ID column values on the integration events list page

Full unique id of the integration event is displayed

Full unique id of the integration event is displayed

Provided validation on endpoint URL length in integration event configuration

Provided validation on endpoint URL length

Provided validation on endpoint URL length

If the endpoint URL, entered by the user, is more than 255 characters an error message is displayed, “… length exceeded the limit”, while configuring an integration.

Availability

The Availability is set as “UP”, however, the device in scheduled maintenance is showing as DOWN

Availability is set as “UP”. Device in scheduled maintenance is showing as Down

Availability is set as “UP”. Device in scheduled maintenance is showing as Down

Identified 2 cases where unable to process availability event:

  1. During Maintenance, resources were in UNMANAGED state and hence could not process the availability event against the UNMANAGED resource.

  2. When the resources are discovered for the first time, availability event is getting failed, as unable to load resource during availability event process; this is due to resource transaction not yet committed.

Now processing availability event against UNMANAGED resources as well.
Sending availability event after committing the resource transaction.

Issue is fixed.

The devices are showing appropriately (showing “UP”) when they are in the schedule maintenance period, when the availability setting is “UP”.