Monitor a VMware NSX Implementation

Configure logging for NSX components according to a deployment plan

 

1: Configure Syslog on NSX Manager

To configure NSX manager to send logs to a centralized syslog server, login to NSX manager UI and click on “Manage Appliance Settings”

Under Syslog server click on Edit button

Punch in your syslog server IP and port 514 and select UDP as protocol and hit OK.

Post configuring syslog on NSX manager, I verified that it is forwarding the logs to syslog manager.

Configure Syslog on NSX Controllers

There is no method available from GUI to set syslog settings on NSX controller and you can only set it via Rest API. The steps of configuring syslog on controllers via Rest API is explained on page 57 of NSX API Guide

1: Get a list of deployed controllers: You can fire below API call to get list of all the deployed controllers

You will get details of all the deployed controllers (if you have more than one). Make a note of the controller-id

Controller’s ID can also be fetched by logging into vCenter and navigating to Networking & Security > Installation & Upgrade tab

2: Set syslog on controller: 

Method : POST

URL : https://NSXMGR-FQDN/api/2.0/vdn/controller/{controllerId}/syslog

Content-Type : application/xml

Request Body

Example: curl -sik -u ‘admin’ -H ‘Content-Type:application/xml’ -X POST https://nsxmgr-01a.corp.local/api/2.0/vdn/controller/controller-6/syslog -d @syslogctlr.xml

where syslogctlr.xml file contains the content mentioned under Request Body

Now if you do a GET against controller-6, you will see that syslog info has been populated there.

Configure Syslog on NSX Edge Services Gateway

To configure syslog on an ESG, double click on the edge and navigate to Manage > Settings > Configuration and select Syslog servers and click on change.

Enter syslog server IP and select UDP as protocol and hit OK.

And syslog status will change to Up

Monitor health of networking services

I am not sure what exactly this objective is about but I guess we can check for

  • EAM status.
  • firewall/VXLAN health status.
  • Hosts preparation status are reporting as prepared. 
  • Communication Channel health status.

Monitor health and status of infrastructure components

Monitoring vSphere Health

Make sure CPU/Memory utilization of your cluster is under threshold. 

Also look for any triggered alarms etc and if found any, then rectify the issue.

Monitor NSX Manager Health

Login to NSX manager UI and from home page verify that all services are running.

Also verify CPU/Memory/Disk utilization has not crossed any configured threshold.

Check for connectivity status of lookup service and vCenter server. Inventory sync should always be recent. 

Check for Audit Logs, System Events and Tasks etc by logging into web client and navigating to Networking and Security > NSX Manager > Monitor tab.

Monitor NSX Control Cluster Health

Controllers general health statistics can be monitored by logging into vCenter and navigating to Networking & Security > Installation Management > NSX Controllers node

To verify  cluster join status and majority status etc, connect the controller node via ssh and run command: show control-cluster status 

To verify the Controller Node’s intra-cluster communication connections status, run command: show control-cluster connections

Enable Data Collection for Single/Multiple Virtual Machines

Will update this section soon.

And that’s it for this post.

I hope you find this post informational. Feel free to share this on social media if it is worth sharing. Be sociable 🙂

Leave a ReplyCancel reply