If you are installing stand-alone Business Automation Workflow on containers or if your
deployment platform is set to the value "other", which is a CNCF platform that is not OpenShift, and
the pfs_configuration.elasticsearch parameter is empty, you must configure the
embedded Elasticsearch. For
21.0.1 Always configure Elasticsearch.
Important: For stand-alone Business Automation Workflow on containers, always
install the latest interim fix for V21.0.2 that includes JR64401 (IF006 or later), or you will get
an error because the operator tries to install the IBM Automation foundation.
Procedure
- Add the following content to the custom resource YAML file:
########################################################################
######## Embedded Elasticsearch configuration ########
########################################################################
elasticsearch_configuration:
## The Elasticsearch pods require the hosting worker nodes to be configured to:
## - disable memory swapping by setting the sysctl value vm.swappiness to 1.
## - increase the limit on the number of open files descriptors for the user running Elasticsearch by setting sysctl value vm.max_map_count to 65,536 or higher.
## When set to true, a privileged init container will execute the appropriate sysctl commands to update the worker node configuration to match Elasticsearch requirements.
## When set to false, you must ask the cluster administrator to change the memory swapping and descriptor properties on each worker node.
privileged: false
## If elasticsearch_configuration.privileged is set to true, you must create a service account that has the privileged SecurityContextConstraint to allow running privileged containers. Refer to Knowledge Center for more info.
## If elasticsearch_configuration.service_account not set, default service account "{{ meta.name }}-elasticsearch-service-account" will be used.
service_account: "<Required>"
- Prepare the security context constraint (SCC) for the embedded
Elasticsearch.