Full read after power outage

[pls]

This issue occurred to all of our backup jobs but I will just use one backup job as an example. The below data is for our exchange server running on vmware. We have an incremental run every 4 hours. We have a vSAN and are using high availability so the total size is double the actual size of the server. The backup at 12am is what a normal backup looks like and takes less than 20 minutes. We had a power outage at about 3:30am on 9/25 and power did not come back up until 7am so all servers would have lost power so the 4am backup was skipped. The 8am backup read the entire server and took over 9 hours. The backups that occur once per day at 6pm had the same behavior. The backups since then have been normal. Any idea why this would have occurred? Is there anything I should have done when the power came back on?
Thanks

Time Total Size Read Transferred
12am 1.2TB 22GB 8.7GB
4am Failed Power Outage
8am 1.2TB 500GB 15GB

[Regnor]

Sounds like CBT wasn't used after the power outage. Did you receive a warning that the CBT data was invalid?

[PetrM]

I'm pretty sure that Regnor's assumption is correct, power outage might cause CBT issues, for example according to this KB:

In some cases, such as a power failure or hard shutdown while virtual machines are powered on, CBT might reset and lose track of incremental changes.

Thanks!