Indy Node has a bug in TAA handling code. The current primary can be crashed with a malformed transaction from a client, which leads to a view change. Repeated rapid view changes have the potential of bringing down the network.
On May 18, Evernym’s monitoring of Sovrin StagingNet showed a report of StagingNet losing sufficient consensus to validate write transactions. The problem resolved itself within a few minutes. On May 20th we saw the alert multiple times, and we began analyzing the logs of our steward node. On May 21st we continued to see the alerts with increasing frequency.
It appears that someone is unknowingly sending a malformed transaction, and retrying when the transaction fails. The cause of the errors appear to be the TAA acceptance.