Restore a cluster to a node with a different IP address
For some time now, we’ve provided a procedure for backing up and restoring your Dynatrace clusters. One of the limitations of this approach is that restore has to happen on a node with the same IP address as the backup node. This requirement has been eliminated with Managed version 1.162 and a new restore procedure has been introduced. To ensure uninterrupted operation, update your local cluster restore procedures before upgrading to OneAgent 1.162.
This new option opens up a whole new set of use cases for creating backup clusters and restoring their data periodically. We’ll describe some of these use cases and best practices for using this new procedure in an upcoming blog post.
Introducing cluster emergency contacts
You can now be notified by the Dynatrace One team of rare and unexpected issues if something serious goes wrong with your Dynatrace cluster. We’ve introduced a new configuration that allows you to specify the contacts who should receive emergency notifications of catastrophic cluster states or critical security vulnerabilities that must be addressed. We do everything we can to avoid situations that might require such alerts, however, you can now register the email addresses of your emergency contacts as a precaution.
Proxy Dynatrace Root Cert download through Mission Control
Taking into account the feedback we’ve received from customers, we’ve made it even easier for you to get started using Dynatrace Managed. In the past, users had to open their firewalls to allow communication from
ca.dynatrace.com to verify the signature of the Dynatrace Managed installer—just to complete the installation. Now, Mission Control is able to proxy this request at installation time by default and no additional configuration is required to follow the installation steps provided in your Managed invitation email.
- Manual upgrade on Managed cluster node triggers rolling upgrade of the cluster
- Elastic and Cassandra data directory flags are ignored during Managed installation
- If an external proxy is modified in
config.propertiesMC connection tests displayed in UI may break
- If SSO is turned on and group assignment is off, we should allow for creating/updating users via the UI and REST API
- If LDAP is turned on, we should not allow the creation and updating of users via the REST API
- No restart of Dynatrace components after timeout on file ownership
- Upgrade to Managed 1.158 failed on iptables
- Managed node stuck in an upgrade state – not possible to start Elastic
- Very long OneAgent download causes upgrade to timeout
- Re-sending invitation to community sometimes doesn’t work
- Dynatrace managed unable to remove a node – failing to reconfigure ES on already removed node
- CMC Node CPU% not displayed
- Host Unit Consumption Incorrect
- Clarified overages label in case overages limit is reached
- Corrected licensing behavior for intermittent and unstable agent connections
Also in this release
- We’ve visually improved our email communication so that it looks cleaner and is easier to consume.
- We’ve decreased the frequency of alerting via cluster events of upgrade download failures. Now one event will be created following 5 upgrade download failures.