Applying regular upgrades to your Nuage Networks deployment is an important part of the SDN lifecycle for your network. Upgrades provide you with the latest in features, bug fixes, and security enhancements.
Before you execute an upgrade, it is important to make sure your network is healthy. MetroÆ includes steps to verify that the Nuage Network components are healthy before and after the upgrade. This helps to ensure a trouble-free upgrade process. During the upgrade, if MetroÆ flags an error, chances are there is some kind of configuration error that needs to be addressed prior to attempting the upgrade. The details of such errors are written to disk on the MetroÆ server. The default location is in the local reports directory, but the location can be customized to your preferred location. If an error occurs, check the contents of reports directory first.
Note: the reports directory contains more than errors. It contains important logs of the configuration of your deployment.
Here is a sample of the health-related operations MetroÆ will execute:
- Verify that all expected programs and services are operating properly on the VSD (monit status)
- Write out the network configuration of all VSD nodes
- Write out the ejabberd cluster configuration of all VSD nodes
- Verify that VSD-VSC connectivity is healthy
- Verify that the VSD REST and JMS interfaces are reachable
- Write out the VSD BGP information
- Optionally verify the following information for the VSC
- Number of BGP peers
- Number of connected vswitches
- Number of connected VM ports
- Number of gateway ports
- Write out the network configuration of ElasticSearch nodes
- Verify that the HTTP interface of each ElasticSearch node is reachable
This is but a small sample of the checks that are done prior to an upgrade.