Last week VMware released NSX-v 6.3.5 (Build 7119875) that contains a few new features and addresses a number of bug fixes from previous releases. Going through the release notes there are a lot of known issues that should be known and there are more than a few that apply to service providers…specifically there are a lot around Logical and Edge Routing functions. The other interesting point to highlight about this release is that this is apparently the same build that runs on VMware on AWS instances as mentioned by Ray Budavari.

The new features in this build are:

  • For vCenter 6.5 and later, Guest Introspection VM’s, on deployment, will be named Guest Introspection (XX.XX.XX.XX), where XX.XX.XX.XX is the IPv4 address of the host on which the GI machine resides. This occurs during the initial deployment of GI.
  • Guest Introspection service VM will now ignore network events sent by guest VMs unless Identify Firewall or Endpoint Monitoring is enabled
  • You can also modify the threshold for CPU and memory usage system events with this API: PUT /api/2.0/endpointsecurity/usvmstats/usvmhealththresholds
  • Serviceability enhancements to L2 VPN including
    • Changing and/or enabling logging on the fly, without a process restart
    • Enhanced logging
    • Tunnel state and statistics
    • CLI enhancements
    • Events for tunnel status changes
  • Forwarded syslog messages now include additional details previously only visible on the vSphere Web Client
  • Host prep now has troubleshooting enhancements, including additional information for “not ready” errors

That last new feature above is seen below…you can see the EAM Status message just below the NSX Manager IP which is a nice touch given the issues that can happen if EAM is down.

If you click on the Not Ready Installation Status you now get a more detailed report of what could be wrong and suggestions of how to resolve.

Important Fixes :

  • VMs migrated from 6.0.x can cause host PSOD When upgrading a cluster from 6.0.x to 6.2.3-6.2.8 or 6.3.x, the VM state exported can be corrupted and cause the receiving host to PSOD
  • “Upgrade Available” link not shown if cluster has an alarm. Users are not be able to push the new service spec to EAM because the link is missing and the service will not be upgraded
  • NSX Manager crashes with high NSX Manager CPU NSX Manager has an OOM (out of memory) error and continuously restarts
  • NSX Controller memory increases with hardware VTEP configuration causing high CPU usage A controller process memory increase is seen with hardware VTEP configurations running for few days. The memory increase causes high CPU usage that lasts for some time (minutes) while the controller recovers the memory. During this time the data path is affected
  • Translated IPs are not getting added to vNIC filters which is causing Distributed Firewall to drop traffic When new VMs are deployed, the vNIC filters do not get updated with the right set of IPs causing Distributed Firewall to block the traffic.

Those with the correct entitlements can download NSX-v 6.3.5 here.

References:

https://docs.vmware.com/en/VMware-NSX-for-vSphere/6.3/rn/releasenotes_nsx_vsphere_635.html