In vRealize Operations product UI, metric collected from NSX-T is not evenly distributed.
In the logs, similar errors to below can be seen:
2021-02-19T13:28:22,151 ERROR [NSX-T Resource Collector thread 10, time: Fri Feb 19 13:25:56 KST 2021] (2477) com.vmware.adapter3.nsxt.resource.TransportNodeResourceCollector.listTransportNodesState - Error in fetching transport node states, exception caused is: com.vmware.nsxt.client.ApiException: at com.vmware.nsxt.client.ApiClient.handleResponse(ApiClient.java:1074) ~[nsxt-client-sdk-2.4.0.jar:?] at com.vmware.nsxt.client.ApiClient.execute(ApiClient.java:990) ~[nsxt-client-sdk-2.4.0.jar:?] at com.vmware.nsxt.client.api.NetworkTransportTransportNodesApi.listTransportNodesByStateWithDeploymentInfoWithHttpInfo(NetworkTransportTransportNodesApi.java:2256) ~[nsxt-client-sdk-2.4.0.jar:?] at com.vmware.nsxt.client.api.NetworkTransportTransportNodesApi.listTransportNodesByStateWithDeploymentInfo(NetworkTransportTransportNodesApi.java:2240) ~[nsxt-client-sdk-2.4.0.jar:?] at com.vmware.adapter3.nsxt.resource.TransportNodeResourceCollector.listTransportNodesState(TransportNodeResourceCollector.java:146) [NSXTAdapter3.jar:?] at com.vmware.adapter3.nsxt.resource.TransportNodeResourceCollector.collect(TransportNodeResourceCollector.java:52) [NSXTAdapter3.jar:?] at com.vmware.adapter3.nsxt.resource.ResourceCollectorTask.run(ResourceCollectorTask.java:21) [NSXTAdapter3.jar:?] at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511) [?:1.8.0_241] at java.util.concurrent.FutureTask.run(FutureTask.java:266) [?:1.8.0_241] at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) [?:1.8.0_241] at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) [?:1.8.0_241] at java.lang.Thread.run(Thread.java:748) [?:1.8.0_241]
Note: The preceding log excerpts are only examples. Date, time, and environmental variables may vary depending on your environment.
There is no such error for vSphere metrics.
Environment
VMware NSX-T Data Center VMware vRealize Operations 8.1.x
Cause
Due to the collection interval variation, such metrics collection variation can be observed.
Resolution
The average expected collection cycle is 10-15 minutes, varying due to the response time from NSX-T APIs.
It is recommended to increase the collection cycle duration in the deployed environment to 10 or 15 minutes.
Apart from the above, we can also change the API_COLLECTION_WAIT_TIME_MINUTE in the NSXT.properties file from 30000(30 sec) to 40000(40 sec). This is not compulsory. This can be altered if variations can still be seen in the metric collection after changing the collection duration.