oi_connector (DX Operational Intelligence Gateway) Release Notes

The DX Operational Intelligence Gateway (oi_connector) probe integrates DX Unified Infrastructure Management (UIM) and DX Operational Intelligence (DOI). You can configure the oi_connector probe to send data to DX Operational Intelligence.
The DX Operational Intelligence Gateway (oi_connector) probe integrates DX Unified Infrastructure Management (UIM) and DX Operational Intelligence (DOI). You can configure the oi_connector probe to send data to DX Operational Intelligence.
The DX Digital Operational Intelligence Gateway (oi_connector) achieves the following:
  • Store UIM alarms and metrics (QoS) in DX Operational Intelligence
  • Build Dashboards with the alarms and QoS information
The probe is configurable only through Admin Console and performs the following tasks:
  • Creates different queues for alarms and QoS on the hub
  • Subscribes to these queues
  • Posts alarms and QoS to Jarvis (organizations or tenants register their analytical applications with Jarvis for using it as the analytics engine)
For the DX Unified Infrastructure Management (UIM) - DX Operational Intelligence (DOI) integration and the supported connector probes, refer the DX Operational Intelligence Interoperability page.
Revision History
This section describes the history of the revisions for the oi_connector probe.
What's New:
  • Added the support to post health data to the DX Operational Intelligence through the connector, including the state of the product, the posted alarms, and the metrics statistics.
November 2021
What's New:
  • Added the Alarm Reconciliation mechanism between UIM and DX Operational Intelligence.
  • Jarvis Ingestion Endpoint and NASS Url are combined into a URL. URL is a gateway.
  • Elasticsearch is deprecated from the oi_connector configuration as Jarvis Elasticsearch URL is not relevant to the later versions of DX Operational Intelligence On-prem 20.2.
  • Added deployment compatibility on Secondary Hubs and Robots connected to the Primary Hub.
  • Added support for getting DX Operational Intelligence alerts by ranges if crossing ES limitation.
  • Removed UIM Groups subscription from the UI.
Fixed Defects:
  • Fixed an issue with the invalid data in the oi_connector reports coming from the url_response probe resulting in invalid SNOW incidents.
    Support Case 32822441
October 2021
What's New:
  • Added direct links to Operator Console for alarms and groups.
  • User Interface and usability improvements.
  • Added a feature to filter the alarms by alarm severity
February 2021
What's New:
  • Enhancements in Capacity Analytics use cases support.
  • Manual cache cleanup not required after tenant details change in configuration.
  • Subscribing to UIM Inventory and Spectrum configurations are no longer supported by the OI Connector. Inventory topology data about the devices from UIM is sent through apm_bridge probe and Spectrum data by Spectrum Publisher.
  • Metric metadata publishing to Jarvis is no longer supported.
  • After upgrading to 1.38 version, the older metric chart would not be available for some probes such as net_connect, rsp, cdm and so on as new metrics will get added to these probes.
  • Performance improvements and defect fixes.
  • User Interface and usability improvements.
September 2020
What's New:
  • Performance Improvements
Fixed Defects:
  • DE444303 - OI Connector does not sync up the deleted devices in Elastic index.
  • DE446071 - Few alarms sent to OI from UIM without cs_id.
  • DE442586 - Few metrics sent to OI from UIM without metric_id. Support Case 20144876.
  • DE469044 - Support updating source/host for alarms.
July 2020
What's New:
  • Added NASS support in the probe.
  • The probe registers UIM metrics metadata to NASS and metrics values are posted to NASS.
  • The probe sends UIM metrics metadata to Jarvis.
  • UIM alarm index is populated with new key “metric_unique_id”.
  • Added support for Elastic Search 7.5.1
May 2020
What's New:
  • The probe now includes the interval value along with every QoS that UIM sends to OI.
November 2019
What's New:
  • Introduced incremental synchronization for device inventories. Incremental synchronization is a mechanism for synchronizing only the device or group's configuration data, without requiring a full synchronizing for each configuration change. Using this feature can reduce the device sync/load time. For more information on configuring this feature, see the
    key in the
    Configure Advanced Properties
    in oi_connector AC Configuration.
  • You can now use a proxy connection (HTTP/HTTPS) to send data from on-premises UIM to OI on DX SaaS, which increases security. For more information, see
    Proxy Configuration
    section in oi_connector AC Configuration.
  • You can now use custom user tags in device inventories. For example, you could use User Tag 1 for geographic location and User Tag 2 for type of server. For some systems, you enter Europe for User Tag 1 in the robot configuration, and for some systems you enter Asia. For User Tag 2, you enter either production server or web server. For more information, see User Tags.
  • Introduced the
    key to enable tenant-specific data ingestion of all UIM data (alarms, QoS, devices, and CIs) to UIM SaaS through the raw configuration interface. If you enable this key, then multi-tenancy configuration is not required. For more information, see
    Configure Advanced Properties
    in oi_connector AC Configuration.
May 2019
What's New:
  • The initial release of the probe.
January 2019
Probe Specific Hardware Requirements
The RAM and CPU cores required for the robot hosting the oi_connector probe depends on the following factors
  • The scale of the environment for open alarms, QOS, inventory, and configuration items;
  • Any advanced features like alarm reconciliation are enabled or not;
    When alarm reconciliation is enabled, oi_connector requires more RAM compared to the regular operations. The memory and threads consumed by the alarm reconciliation operation also depend on the time duration/interval of reconciliation. Reconciliation with a shorter duration consumes less memory and processing threads.
  • Any other high-scale probes deployed on the robot.
For the installation of the oi_connector probe, follow the UIM Sizing Recommendations.
resources requirements are:
  • Memory: 6 GB of RAM (+2 GB for the reconciliation functionality using); The OOB configuration of the probe requires 512 MB of RAM
  • CPU: 2.8-GHz fore-core processor 64-bit
Probe Specific Software Requirements
The oi_connector probe requires the following software environment:
  • CA Unified Infrastructure Management 9.0.2 or later
  • Robot 7.80 or later (recommended)
  • Java JRE version 8 or later
  • DX Operational Intelligence 1.3 or later
  • ci_defn_pack version 1.22 or later updates the DX UIM database with new metric information. This package is required where the nis_server probe is running. Restart the nis_server probe if you update this package.
Before deploying the oi_connector probe on your robot, ensure that the following pre-configuration requirements are met:
  • DX Operational Intelligence version 1.3 or later must be installed and all the required mappings and indices must be on-boarded. For more information, see Add DX UIM.
Before configuring multi-tenancy for the oi_connector probe, ensure that the following pre-configuration requirement is met:
  • Webservices_rest
    probe version 8.51 or later must be installed on your system where the UMP/OC is installed.
Installation Considerations
Consider the following before you install the probe:
  • To avoid vertical scaling of the primary hub machine, install the oi_connector probe on the robots connected to the primary hub (including the secondary hub robots connected to the primary hub). Co-locate the robot hosting the oi_connector probe and the primary hub to avoid network latency.
The probe maintains queues for alarm and QoS and if these queues contain old alarm and QoS, the probe sends this information along with the new QoS and alarms.
  • Secure Hub & Robot (CA UIM 9 SP1 and above) - In a secure environment, if any probe that is installed on an independent secure robot tries to subscribe to queues in the related secure hub, the probe fails to attach to the queues. As a workaround, if the probes require to read or publish to a hub queue, then deploy the probe on the primary hub robot. For more information, see Secure Hub and Robot.
Upgrade Considerations
(version 1.41 or later) Upgrade from older versions of oi_connector:
The probe 1.41 version uses an encrypted token. While upgrading from the older versions, the probe verifies and encrypts the token. In case of downgrading to an older version, you must reconfigure the token using the AC profile.
(version 1.38 or later) Upgrade from older versions of oi_connector:
While upgrading to oi_connector 1.38 or later versions, check and update the key
value to '
' in the oi_connector.cfg file. The key
can be found in the
section of the oi_connector.cfg file.
(version 1.32) Upgrade from older versions of oi_connector:
UIM metrics/QoS metadata and values are posted to NASS by default with version 1.32 upgrade. If you want to send the data to Jarvis only, then follow the below steps:
  1. Deactivate the oi_connector probe.
  2. Go to raw configuration.
  3. Go to setup section and edit the value of
    key. Set the value to false.
  4. Save and Activate the probe.
In the probe version 1.32, the default value for the
key doc_type_version_uim_metrics_metadata
is set to "2". For the oi_connector probe connected to a SaaS application, this key has to be manually updated to "2" as upgrade will retain the previous value.
(version 1.20) Upgrade from axagateway to oi_connector
Follow these steps:
  1. Uninstall axagateway.
  2. Download and install the latest version of oi_connector.
  3. Configure the oi_connector from using the Administration Console.
Connector Health Monitoring
Oi_connector probe posts health data of the connector to the DX Operational Intelligence along with the state of the product, the posted alarms, and the metrics statistics. Using the Connector Health UI of the DX Operational Intelligence, you can get an overview of the health and key metrics of the connectors (oi_connectors) present in your deployment. You can monitor the state and utilization of these connectors. You can avoid any unexpected connector issues by monitoring the health and key metrics of the connectors. For more information, refer Connectors Health Dashboard.
Each UIM connector is set up with its own independent identity in the Connector Health feature space.
  1. After making any changes in the probe configuration, it sends a structure as a vertex in TAS and is linked to the respective connector. Each connector sends the following metrics to TAS using TAS API in the specified time interval:
    For the connector’s product-level oi_connector posts:
    • Product Name
    • Product Details
    • Product Version
    • Product Status
    For the connector level oi_connector posts:
    • Unique ID 
    • Connector Name
    • Connector Version 
    • IP Address
    • Host
    • Status
    • IngestionStatus
    • Timestamp
  2. Each connector sends the following metrics to NASS using NASS API in the specified time interval:
    • Unique ID
    • Number of metrics
    • Number of alarms
    • Timestamp
  3. (Optional) Change the TTL for the Product and the Connector Vertex.
    By default, the
    for the Product and the vertex is set to 4 days. If you want to change this, make an entry in the oi_connector probe configuration file.
    vertex_ttl_in_sec = <Value in sec>
  4. (Optional) Change the connector health data-posting interval.
    You can change the connector health data-posting interval in the
    of the probe configuration.
    Default Value: 5
For the highly scalable and loaded systems, the oi_connector health functionality supports a multi-connector mode: the possibility to post connector data for several oi_connector probes instances that are configured on separate hubs robots with different independent queues. To avoid the connectors’ information overlapping, you much provide each connector instance its own subsequent number. You can do it by configuring the
configuration key.
You can verify the data ingestion into the connector health dashboard by the following steps:
  1. Configure the connector to send monitoring data to the Dashboard.
  2. Open
    Connector Health
    dashboard to validate the following:
    • The status of the connector
    • Details of connector like version, IP, name, etc.
    • Details on ingested health metrics
Known Issues and Workarounds
This section describes the known issues and workarounds for the oi_connector probe.
Probe GUI does not open or does not respond
Sometimes the probe GUI might not open or might not respond due to incorrect access rights to the probe. To fix this, you must update the current access rights of the probe.
Follow these steps:
  1. Open Infrastructure Manager.
  2. Select the
    probe, right-click, and select
    Set Access
  3. Provide Administrator or SuperUser access as required.