24 Jul 17:06

3792669

marklogic-2.0.0 Latest

Latest

Release 2.0.0

With version 2.0.0 of Helm Charts, MarkLogic Rootless images are supported and compared to previous Helm Charts versions, PODs are created in parallel instead of sequential which allows faster creation or update of the MarkLogic Clusters. Helm Charts 2.0.0 version also supports Path Based Routing with Ingress capability. Refer to Kubernetes-Path-Based-Routing to enable the fields required to use this feature.

Use of Legacy Host Names

Starting with Helm Charts Version 1.1.2, Stateful sets naming convention got changed which caused upgrade from 1.0.x to 1.1.2. A new parameter called useLegacyHostnames was added to the values file to allow upgrade from version 1.0.x. It defaults to false but when upgrading from version 1.0.x to 1.1.2 or later, this parameter must be set to true.

If users want to use the shorter host names and avoid the usage of legacy host names starting with Helm Version 2.0.0 then data should be backed up first then existing Statefulset should be deleted and new Statefulset should be created with 2.0.0 versions. Once the cluster is up and running restore the data.

Helm Charts Version Upgrade to 2.0.0

MarkLogic recommendation is always to have backup of the data before any upgrade is performed. Refers to the documentation link for more information related to backup and restore Backup and restore a database

Delete the stateful set created using previous Helm Chart version:

kubectl get sts -n <release-namespace> (note down the statefulset name)
kubectl delete sts <statefulset-name> -n <release-namespace>

If upgrade is from Helm Charts version 1.0.x set useLegacyHostnames to true. Refer to Helm Charts 1.1.2 release notes and also set the allowLongHostnames to true if hostnames length is greater than 64 characters

Run the Helm upgrade command. Specify the name of your release and the new chart version using the --version option. Specify the values.yaml file using the -f option:
helm upgrade <your release> marklogic/marklogic -f values.yaml --version <new version> -n <release-namespace>

Verify StatefulSet and Pods are created successfully

kubectl get sts -n <release-namespace>
kubectl get pods -n <release-namespace>

Assets 3

29 May 21:20

github-actions

marklogic-1.1.2

c2bb142

marklogic-1.1.2

Release 1.1.2

Version 1.1.2 of the Helm chart for MarkLogic Server fixes a bug that prevents an upgrade from 1.0.x to 1.1.1 because of the StatefulSet naming convention change introduced in version 1.1.1.

A new parameter called useLegacyHostnames was added to the values file to allow upgrade from version 1.0.x. It defaults to false but when upgrading from version 1.0.x to 1.1.2 or later, this parameter must be set to true.

Because there are limitations to the length of fully-qualified hostnames in certificates, version 1.1.2 of the Helm chart provides feedback if the configured hostnames are more than 64 characters. If the generated hostnames are greater than 64 characters an error will be thrown and the upgrade will fail.

A new parameter called allowLongHostnames was added to the values file to allow overriding this behavior. This defaults to false but can be set to true if hostnames greater than 64 characters are acceptable for the target deployment.

Assets 3

27 Mar 17:46

github-actions

marklogic-1.1.1

1bdaa2f

marklogic-1.1.1

Release 1.1.1

WARNING: Because the StatefulSet naming convention has changed in 1.1, upgrading from version 1.0 of the MarkLogic Helm chart to 1.1 will cause the name of the StatefulSet to change and therefore, the original StatefulSet will be deleted and a new one will be created. This is equivalent to deploying a completely new MarkLogic cluster. The original persistent volumes will NOT be deleted so any data associated with the original MarkLogic StatefulSet can be recovered from the original persistent volumes if needed.

UPGRADING: To upgrade an existing MarkLogic cluster created from version 1.0 of the Helm chart, backup the existing cluster, upgrade the Helm chart to 1.1, and restore the backup to the new cluster. Once the system has been verified, the old persistent volumes can be deleted.

MarkLogic Server is a multi-model database that has both NoSQL and trusted enterprise data management capabilities.

Patch release 1.1.1 of the MarkLogic Helm Chart.

Bugfixes:

Fix for updating the existing cluster from HTTP to HTTPS.
Addressed the issue causing bootstrap host to become unresponsive in PodInitializing state post-restart.

Assets 3

27 Feb 23:52

github-actions

marklogic-1.1.0

67cfd7c

marklogic-1.1.0

Release 1.1.0

This minor release provides the following enhancements:

Default App Servers can now be configured with HTTPS.
Add Pod Topology Spread Constraints feature.
HugesPages can now be configured in the Marklogic pods' resources.
LoadBalancer HAProxy is now using the LTS release HAProxy 2.9.4.
FluentBit is now using the version 2.2.2.
The Helm chart now comes with the latest MarkLogic version available 11.1.0.

Assets 3

19 Dec 17:43

github-actions

marklogic-1.0.2

dda0b59

marklogic-1.0.2

Patch release 1.0.2 of the MarkLogic Helm Chart.

Bugfixes:

MarkLogic containers will never be marked as ready when the Security DB is not available
Special character '$' not accepted for admin password when used as Docker secrets
Additional volumes are not created with the right owner
Update 3rd party images to resolve severity vulnerabilities

Assets 3

09 May 21:33

github-actions

marklogic-1.0.1

e9621b4

marklogic-1.0.1

Patch release 1.0.1 of the MarkLogic Helm Chart.

Bugfixes:

Fixed a race condition in the post-start hook that caused the pods to fail to start when running on macOS on Apple silicon.

Assets 3

25 Apr 02:19

github-actions

marklogic-1.0.0

29cd949

marklogic-1.0.0

This is the first release of the MarkLogic Helm Chart for Kubernetes. With this chart, MarkLogic clusters can easily be created and managed in Kubernetes.

See the MarkLogic on Kubernetes documentation for complete details.

Assets 3