This message was deleted.

Can you share the MM log from the failed MM. Is this on kubernetes? k8s kills pods if the start using more memory than they are allotted. It could be configuration of the Peons and MMs. What is your MM config and node/container characteristics in terms of CPU/memory?