Dynatrace Managed feature update, version 1.168

Updated sliding time-frame window for custom metrics

Until now we had a sliding window of 28 days for custom metrics, which meant that you could consume unique metrics up to your licensed quota within a time frame of 28 days. This also meant that you had to wait 28 days until you could start collecting new metrics. We’ve now reduced the sliding time frame window to 24 hours. This enables you to collect more unique metrics within a shorter period of time without increasing your overall licensed quota.

Revised cleanup strategy for Elasticsearch snapshots

Previously, we retained hourly snapshots of Elasticsearch for 24 hours, daily snapshots for a week, and weekly snapshots for a month. Monthly snapshots were retained indefinitely for backup purposes. However, this approach eventually led to storage issues. Older snapshots don’t contribute much value anyway because RUM data is kept only for a maximum of 35 days and problem history is retained indefinitely regardless. To address this storage issue for older snapshots, we’ve adapted our cleanup strategy so that monthly snapshots are now removed automatically after 6 months.

Also in this release

  • Until now, the Dynatrace Managed installer displayed a warning for installations that have more than 6 cluster nodes. We now support up to 12 nodes per cluster.
  • The JRE of all Dynatrace Managed components was updated to Oracle Java 8u202.

Resolved issues

  • Synthetic tests in CMC endpoint configuration didn’t work when no port was set. In such cases, a default port was assumed which didn’t work in load balancer scenarios.
  • An error page was presented when a user tried to log in to an environment with insufficient permissions.
  • The backup configuration screen now points users to the correct Backup/Restore help page.
  • Users are logged out and blocked after 3 subsequent invalid attempts to change the password on the User profile page.
  • Timeout to determine upgrade status has increased from 30 to 60 seconds.
  • Incorrect “Node down” notification was fixed and an additional alert is now displayed when the node is back up.
  • Event Log now contains successful login entries when users logged in via SSO.
  • Configuration traffic per cluster node wasn’t preserved during upgrades.
  • Problems with custom datastore location and server file permissions during restore from backup have been fixed.
  • An issue around clusters initially downloading the latest major version of OneAgent and all minor builds has been resolved.
  • The emergency email address field in CMC is now empty by default.
  • Updating the email test address on the SMTP Server configuration page no longer triggers the confirmation panel.
  • Changing the license also changed process group names within self-monitoring environments. The cluster identifier is now used instead.
  • Sometimes DEM was disabled for environments even though the usage limit wasn’t met.
  • We fixed the problems with disabling WebUI nodes.
  • Support archive generation in CMC with all switches turned off created an archive with incorrect contents.
  • The misleading “Server left intact” error message has been removed when uploading idp.xml for SAML configuration.
  • SMTP server validation failed when the name contained an underscore character.
  • OpenLDAP without memberOf support matched users only by usernames. Now it matches also by user DN.
  • The cluster’s user management REST API caused inconsistencies due to the delay in synchronizing user/group data across cluster nodes.
  • Adding several users via the REST API worked but resulted in an Error: Not acceptable message if a user already existed.

Other new features

Additionally, all new features introduced with Dynatrace SaaS version 1.167 and version 1.168 are now also supported by Dynatrace Managed.