Released: 2023-04-12
Kindly note that our release process is changing. In future, we will release updates once per month (in the last week of each month). This change increases efficiency for the development teams and should not significantly influence delivery times for feature requests. In the current process, feature requests are groomed in 1 sprint, then implemented in the following sprint. In the new process, the grooming and implementation can be executed within a single sprint.
Note: Additional releases are possible in case of any critical changes or urgent feature updates.
See compared diagram below
We added the Step Name column next to the Flow Name on the Logs page. This makes it easier to isolate relevant logs during trouble shooting.
Note: While hovering on
Mappericon the tooltip will display information with steps interaction.
We applied a new design to the History tab on the Flow-Designer page and transfer it from ember to react
Special Release Procedure (if haproxy is exposed)
Firstly deploy haproxy, and only then the platform. Otherwise, some services will crash.
Helm config changes:
.Values.global.variables.SECONDARY_PLATFORM_CLUSTERS to admiral chart in the next format
SECONDARY_PLATFORM_CLUSTERS:
- name: 'secondary-cluster' # required
credentials:
name: 'my-user' # required
password: 'some-password' # required
disabled: false
disabledRealtimeFlows: false
.Values.global.services..enabled (default: true)service.beta.kubernetes.io/azure-load-balancer-ipv4 to all internal load balancers if cloud provider is AZUREvalues-secondary.yaml file to admiral and platform charts for disabling services that do NOT run on secondary clustersdeploy.sh to deploy the platform to secondary cluster (i.e. using values-secondary.yaml). The command example:KUBE_CONTEXT=great-moraq-stannis USE_SECONDARY=true ./deploy.sh deploy great-moraq
Changes required in configurations in k8s clusters (e.g. stage-clusters unmerged PR):
haproxy.yaml add expose: true and loadBalancerIp: ""In platoform.yaml:
2.1. In global.services.api add expose: true and loadBalancerIp: ""
2.2. In global.services.maester add expose: true and loadBalancerIp: ""
2.3. In global.services.steward add expose: true and loadBalancerIp: ""
KUBE_CONTEXT=great-moraq-stannis USE_SECONDARY=true ./deploy.sh deploy great-moraq
Changes required in configurations in k8s clusters (e.g. stage-clusters unmerged PR):
haproxy.yaml add expose: true and loadBalancerIp: ""In platoform.yaml:
2.1. In global.services.api add expose: true and loadBalancerIp: ""
2.2. In global.services.maester add expose: true and loadBalancerIp: ""
2.3. In global.services.steward add expose: true and loadBalancerIp: ""
2.4. In global.services.pssLoadBalancer add expose: true
secrets.yaml change all occurrences of haproxy-service.platform.svc.cluster.local to an actual haproxy loadbalancer IP address500 TimeoutError on GET v2/logs in case of applying together multi-filtering and search1.4.2 (v23.14)Don't encrypt payload and Content-Type configuration fields to Publish actionDon't decrypt payload configuration field to Consume triggerSailor library version to 2.7.1amqplib library version to 0.10.3Publish actionConsume trigger1.4.4 (v23.14)Re-assemble Messages mode1.1.0 (v23.14)Lookup Object (at most one) ActionUpsert Object ActionReleased: 2023-04-21
We have simplified the process of copying the Webhook link. Clicking on the link will no longer open a new browser tab. Instead, the link will be automatically copied to the clipboard.
Released: 2023-05-10
Starting from the next release 22.23 the Recipe feature will be available for the tenant-level of subscription. For Enterprise subscriptions the Recipe feature will be available at additional cost.
Note: If you are interested to use the Recipe feature functionality please contact sales@elastic.io for details.
node_config option consume_raw_data into the Flow’s attributes as a configuration option. This parameter configures the parsing behaviour in the webhook step. If selected, the webhook step receives an input message in the raw structure.
raw in the Send Sample request.Consume Raw Data configuration for step in the Advanced Settings menu.With this feature request GET v2/containers now returns reason and exit_code attributes.
This is a part of the platform optimisation and extending of its capability to withstand a high load. For this reason, we are moving forward to remove the Runlog and use a new “Containers” page that will display related information.
Some instagram services use unusual authorisation schemas, so we added a New Auth Client Type oauth2_instagram.
In the component.json of REST API V2 component you need to define the scopes, because OAuthFieldView does not allow to specify it in credentials. E.g.:
"authClientTypes": [
"oauth2_instagram"
],
"credentials": {
"fields": {
"auth": {
"required": false,
"viewClass": "OAuthFieldView"
}
},
"oauth2": {
"scopes": ["user_profile", "user_media"]
}
},
Note:
HTTPAuthViewdoes NOT supportoauth2_instagramtype.
We applied a new design to the Summary tab on the Flow-Designer page and transfer it from ember to react.
We are informing you that we will be performing a Kubernetes upgrade (v1.26) on our systems on 25/05/2023 from 17:00 to 19:00 (GMT+2). During this time, some of our services may be not be available for short period, and you may experience intermittent connectivity issues. This upgrade is necessary to ensure that we continue to provide the highest level of service to our customers, with improved stability and security. The Kubernetes upgrade will enable us to take advantage of the latest features and security enhancements, which will improve our overall system performance and help us better meet the needs of our customers. We understand that any maintenance downtime can be inconvenient, and we will work to minimize the impact on your business operations. Our team will be monitoring the systems during the upgrade to ensure that everything runs smoothly. If you have any questions or concerns, please do not hesitate to contact our customer support team, who will be available throughout the maintenance period.
PATCH /v2/tenants/{tenant-id}/roles makes changes in Database even if it fails.1.2.11 (v23.19)json-bigint library3.2.0 (v23.19)Read CSV attachment action:
Skip empty linesComment charCreate CSV From Message Stream and Create CSV From JSON Array actions:
New line delimiterEscape formulae1.2.0 (v23.19)Delete Object ActionLookup Objects (plural) ActionGet New and Updated Objects Polling Trigger1.1.1 (v23.19)Sailor library version to 2.7.12.3.0 (v23.19)Get New Leads Polling:
Emit page option to Emit BehaviourPage Size, Max iterations and Return LeadsSailor library version to 2.7.1node v181.2.2 (v23.19)Get New and Updated Objects Polling TriggerReleased: 2023-06-14
The elastic.io platform has been upgraded to support Kubernetes v1.26. Please follow this link to see Kubernetes v1.26 its own changelog.
Released: 2023-06-16
Runlog has been removed from Dashboard page (v23.24)As a part of our ongoing platform optimisation and extending its capability to withstand a high load, we have removed the Runlog and implemented a new “Containers” page that will display related information. This change does not reduce the amount of logging information available to you. Instead it increases the information available for error handling and debugging while increasing the efficiency of the platform.
Containers Page (v23.24)We implemented new menu Containers under the Analyze section.
The new page Containers will display information about containers and components inside.
The main page displays logs for “Today” by default, and users can extend the date range by the filtering and Search fields. The container logs save up to 30 days.
The Containers page displays next States:

To open information for the particular container click on Step Name. The single container page displays next information:

NOTE: Logs collected before this release won’t be present on Container Details page.
As we disabled the data messages processing in the lookout we created a new data source Clickhouse for Dashboard.
containers_id (v23.24)We added new query parameter container_id into GET v2/logs endpoint.
We changed the helm env var RECIPES_FEATURE_DISABLED to true. It means that the Recipes are disabled for all contracts.
Before deploying, Tenant-Admin should update the feature flag for the necessary contracts with disableRecipeFeature: false to continue using the Recipe feature.
See in API DOCS
{
"data": {
"type": "contract",
"id": "contract_id",
"attributes": {
"feature_flags": {
"disable_recipe_feature": false
}
}
}
}
Note: The Recipes are available to activate. Only create, edit and delete are resricted.
The Lookout does not process data/input messages to reduce the Mongodb load.
We introduced the new env var - LOOKOUT_PROCESS_DATA (default: false).
It controls whether the lookout consumes data and input non-retry messages.
The particular solution to fix an issue in CSS styles. With the next release, this section will be transferred to React.
Details of errors that occurred more than 30 days ago have been removed and are no longer accessible.
The amount of errors produced by this Flow exceeds error record limit. Older error records were deleted to persist the most recent ones.
2.0.0 (v23.24)1.2.3 (v23.24)Lookup actions and Get New and Updated Objects Polling trigger1.0.0 (v23.24)Make Raw Request action1.2.0 (v23.24)Get New and Updated Objects Polling trigger1.6.0 (v23.24)Webhook trigger in favor of Hubspot webhook component1.0.0 (v23.24)Webhook trigger1.3.0 (v23.24)Jira Service Management Cloud2.0.1 (v23.24)metaHeaders and searchHeadersUpsert Object action did not save the search headers in MaesterDon't throw an error when updating an object with TTL2.6.0 (v23.24)Don't emit on empty results checkbox in Query trigger1.4.0 (v23.24)attachments inside message body for XML Attachment to JSON action1.2.0 (v23.24)Duplicates behavior optionFrom this table to From this columnTo this table to To this column1.0.4 (v23.24)