← Back to the front page
- It is now possible when using the Native scheduler to specify Docker logging driver configuration for each container using the
- A new HTTP Request Programmable Preflight Check has been added for the Native and Swarm schedulers.
- The Replicated Admin Console browser session will now time out after 30 minutes of inactivity to better align with NIST guidelines. Previously this time out was set to 24 hours.
- The Native and Swarm installation scripts ensure that /etc/replicated.conf is world-readable if it exists so that it can be read by the Replicated daemon.
- Support bundle parallel task execution concurrency is now limited to the number of CPU cores up to a maximum of 4 to prevent starving the system of resources. Previously this limit was set to the number of CPU cores.
replicatedctl snapshots ls command will now return snapshots in date order instead of ID order.
- The statsd container will no longer use host networking when passing the
disable-replicated-host-networking flag to the Replicated Native Scheduler installation scripts.
- Fixed an issue that could cause Replicated to fail to garbage collect unused images from previous versions of the application.
- Fixed an issue that could cause Replicated to show a “no updates to apply” error when attempting to upgrade the application.
- Fixed an issue that could prevent docker 20.10 from installing on RHEL 7. Docker 19.03 will remain the default installed version and the highest supported version.
- All known CVEs with fixes have been patched in all images distributed by Replicated at the time of the release. For more information see this article.