The bandwidth limit for the VMs is only 3Mbps. However, we noticed in the network interface there is spike more than 20 Mbps. What causes this, and how can I resolve it?
There are two possible reasons for this.
There is a known bug during Upgrading that can cause this. Unfortunately we don't have a resolution for this at this time.
If you were not performing an Upgrade during this time then it is far more like that such peaks mean that the CP wasn't receiving stats data for a while and then got all of the summarized stats in one moment.
It is can be related to stop/start of the onapp daemon or cron service, heavy load on network etc.
It doesn't mean that VMs used more network bandwidth than the limit.
This is not affected by backup traffic either, as OnApp backups don't have an effect on the appliance network.