Hi,
In our environment we use VMware as virtualization platform.
On-premise FNMS instance is integrated to number of vCenters to collect inventory details. As we use IBM PVU licenses, IBM PVU sub-capacity tracking with 30 min. scans is enabled in our FNMS environment.
Our VMware environment is configured to use Distributed Resource Scheduler (DRS). The main goal of DRS is to ensure that VMs and their applications are always getting the compute resources. This result in hundreds of VMs being moved between ESXs every hour. We have quite conservative configuration, with reduced number of vMotion migrations.
During last year we spotted number of issues with FNMS inventory gathering and processing for our setup. Couple of examples:
- Active inventory without hosts. It looks like FNMS fail to import/process VM - host relations from vCenter if VM is moved at the inventory gathering moment. So almost daily we can see random machines without hosts.
- Corrupted license consumption reports (IBM PVU license). PVU calculations are separated from rest of inventory processing, so two different processes use different data for calculations. Data might differ because of mentioned constant changes in infrastructure. As result we have inventory hierarchy prepared within inventory import, but consumption calculations made separately. Resulting in reports having VMs or hosts with zero consumption, or hosts without VMs, but consuming PVU points.
These and some other issues reduce trust in reporting correctness.
Flexera support see us as the only customer suffering with similar issues. So wanted to check if anyone are experiencing similar or some other issues when they collect inventory from VMware infrastructure with enabled DRS?
Marius
