during the installation of Dynatrace Managed cluster node, the /etc/hosts is modified (added hostname for the loopback IP address). This, albeit mentioned in the documentation, might break other services on the host. In this case, we need to have the hostname in /etc/hosts in the record for the primary IP address, not for the loopback address.
I see there is an option to disable that behaviour in /etc/dynatrace.conf:
HOSTS_ALTERING_ENABLED = true
Solved! Go to Solution.
Hey @Dave M. do you have any insights into this? or know anyone who might be able to answer this question?
Personally prior to messing around with the Clustered nodes, I always check with support by opening a ticket up or reaching out to Chat.
I got an answer from Dynatrace. This change was introduced because Cassandra could not start up properly on some Linux distributions when host entry was missing in /etc/hosts.
Ahh thanks for sharing your findings with us 🙂