vRealize Operations Management Pack for NSX on vSphere | 24 Aug 2021 | Build 18446371
What's in the Release NotesThe release notes cover the following topics:
- Upgrading from the Management Pack for NSX on vSphere
- Content Pack for Log Insight
- What's New
- Known Issues
The vRealize Operations Management Pack for NSX on vSphere extends the operational management capabilities of the vRealize Operations platform to the areas of virtual and physical data center networking.
Before you upgrade from the Management Pack for NSX on vSphere, perform the following steps:
- Stop all instances of the NSX on vSphere adapter.
- Click Administration and then click other accounts.
- Select adapter instance for NSX-vSphere and click Stop Collecting.
- Upload the PAK file that you downloaded > select both the tick boxes before upgrading > Click Next > Accept the agreement > Next > Finish.
- Start the instance of NSX on vSphere.
- Click Administration and then click other accounts.
- Select adapter instance and click Start Collecting.
Supported Upgrade path for NSX on vSphere
- NSX on vSphere 3.5.2 to NSX on vSphere 3.6
- NSX on vSphere 3.5.2 to NSX on vSphere 3.6.1
- NSX on vSphere 3.6 to NSX on vSphere 3.6.1
- NSX on vSphere 3.6 to NSX on vSphere 3.7
- NSX on vSphere 3.7 to NSX on vSphere 3.7.1 for both On-Premises and SaaS
The Management Pack for NSX on vSphere 3.5.2 is packaged with version 3.6 of the content pack for Log Insight. If you have configured Log Insight with vRealize Operations Manager, the content pack is installed when you install the management pack, enabling integrated workflows between Log Insight and the management pack dashboards. However, automatic installation of the content pack for Log Insight only occurs with Log Insight 3.0 and later versions. For more information, see NSX on vSphere Log Insight Content Pack 3.6 Release Notes.
The Management Pack for NSX on vSphere collects data on NSX objects through NSX APIs. The scale of the management pack depends on the number of objects and metrics collected per object. Depending on the number of virtual machines in the environment and associated network topology requirements, the number of NSX objects varies in each environment. For the NSXobject related scale numbers, see the NSX documentation.
The Management Pack for NSX on vSphere has been tested up to a scale of 2000 NSX objects. However, the management pack continues to work if the object counts exceed 2000 virtual machines and 300 edges. A vRealize Operations Manager cluster with three nodes was used for scalability testing. The cluster consisted of the primary node, a single data node, and a remote collector.
For compatibility between products, please refer to the VMware Product Interoperability Matrices.
- NSX for vSphere Management Pack using basic authentication for Suite-API calls fails with an error message.
To prevent authentication failure, NSX on vSphere Management Pack uses token-based authentication for Suite-API calls between NSX on vSphere Management Pack and vRealize Operations Manager 8.0 or later.
- NSX for vSphere Management Pack version 3.5.2 fails to get controller syslog settings with NSX version 6.4.2 or later.
NSX on vSphere Management Pack version 3.5.2 fails to get controller syslog settings with NSX version 6.4.2 or later. This is due to the incompatibility of the APIs in NSX on vSphere Management Pack 3.5.2 and NSX 6.4.2. The NSX on vSphere Management Pack 3.6.1 supports APIs of NSX version 6.4.2 or later.
- NSX for vSphere version 3.5.2 displays different metrics data for NSX Edge virtual machine memory usage.
NSX for vSphere version 3.5.2 displays different compute (CPU, memory, Disk) metrics of NSX-Edge nodes compared to vCenter virtual machine values. In NSX for vSphere Management Pack 3.6.1 all the metrics data inherited from NSX for vSphere Edge is the same as that of vCenter virtual machines.
- Relationship between NSX-v Objects and vCenter Objects broken in the SaaS Environment
The relationship between NSX-v Objects and vCenter Objects was broken.
This issue has been fixed.
The traceflow action is not currently supported between two virtual machines that reside in different vCenter servers.
The traceflow action may fail when it is run between virtual machines that have more than one virtual NIC.
If Log Insight is configured after the Management Pack for NSX for vSphere is installed, the content pack for NSX for vSphere is not installed automatically.
To install the content pack manually
- Edit the configuration of the Management Pack for NSX for vSphere adapter instance in vRealize Operations Manager.
- Click Test.
The content pack is deployed on Log Insight.
In some environments, the Management Pack for NSX for vSphere impacts the performance of NSX edges because of the volume of API calls.
Reduce the number of API calls by disabling the collection of all metrics and properties for the following object types:
- DHCP Edge Service
- IPSec VPN Edge Service
- Load Balancer Edge Service
- SSL VPN Edge Service
The Check for synchronization issues and the Check host configuration actions cannot be executed on universal logical switches that span multiple vCenter servers.
When using the Management Pack for NSX for vSphere in a multi-vCenter NSX setup, the NSX for vSphere Universe dashboard does not display new NSX for vSphere Managers that are added to the environment.
- Verify that the NSX-vSphere Universe dashboard is displayed on the Home page. If it is not displayed, perform the following steps:
- Click the Content icon and select NSX-vSphere Universe.
- Click the Configure icon and select Add Dashboard(s) to Home.
- Select Environment > Custom Groups and expand Environment.
- Select NSX-vSphere Primary Environments, and from the Actions menu, click Edit Group.
- Select the Keep group membership up to date check box if it is not selected already.
When you run the Show runtime state for host action for a Universal Logical Router, you cannot see all the hosts from each vCenter server. Currently only hosts from a single vCenter server are displayed.
Do not manually uninstall any instance of the content pack for Log Insight before installing the Management Pack for NSX for vSphere. This creates multiple definitions of the same alert, and specific alerts are not triggered. The Management Pack for NSX for vSphere installs the content pack for Log Insight that is optimal for the NSX environment and overwrites any other version.
After upgrading from the Management Pack for NSX for vSphere 2.0, reconfigure the Network Device adapter and add a new credential.
The Universal Synchronization service is not running alert does not trigger if the NSX Management service is down.
When configuring the adapter instance, testing the connection on the NSX Manager fails if the NSX Manager’s certificate uses the RSASSA-PSS encryption algorithm. This algorithm is not supported by vRealize Operations Manager. Hence, you must use a different certificate.
When using Log Insight with vRealize Operations Manager, alerts from the Management Pack for NSX for vSphere do not appear as enabled in the Log Insight Admin console. Alerts are still forwarded to vRealize Operations Manager.
The Management Address property of hosts is only available if the host is running ESX 5 or later.
Collection of network statistics on Edge firewalls is disabled by default in order to reduce the load on NSX and optimize the collection time of the management pack in large environments.
To enable the network statistics for collection in the active policy, perform the following steps:
- On the Administration tab, click Policies.
- On the Policy Library tab, edit the current active policy. Default Policy is selected by default. Click Edit Selected Policy.
- In Override Attributes, select NSX-vSphere Edge Firewall Service as the Object Type.
- Enable the Network|Connections metric.
- In Override Analysis Settings, select NSX-vSphere Edge Firewall Service as the Object Type.
- Enable the Workload section and check the Network entry under Resource Container.
- Enable the Capacity Remaining section and check the Network entry under Resource Container.
- Enable the Stress section and check the Network entry under Resource Container.
- Click Save.
If one or more hosts or virtual machines stop collecting data from vCenter, the Management Pack for NSX for vSphere continues to show those objects as collecting data.
When you run a VXLAN Tunnel Endpoint Connectivity test on a logical switch, the test fails if the NSX for vSphere adapter instance name is identical to the name of any object that has been collected by vRealize Operations Manager, for example, NSX Manager, or Controller Cluster.
When you configure the NSX for vSphere adapter, ensure that you give the adapter a unique name.
If no traffic is flowing through the virtual servers of the NSX-vSphere Edge Load balancer services, the traffic statistics of the Load balancer pool members is not collected, and pool members are not displayed in the metric picker.
Enable traffic on the NSX-vSphere Edge Load balancer virtual servers.
The Logical Router is deployed to the same host as one or more ECMP Edges alert is triggered when ECMP is enabled on one of the edges. This issue occurs when one or more ECMP edges do not get cleared after disabling ECMP on the edge.
Clear the ECMP edges when ECMP is not enabled on the edge.
During the upgrade of vRealize Operations Manager, some icons of the Management Pack of the NSX for vSphere disappear. This affects the dashboard and widgets of the Management Pack for NSX for vSphere.
Reinstall the Management Pack for NSX for vSphere.