After upgrading to a new platform version, always ensure that your installed component versions match the versions specified in the release notes. To verify this, navigate to FlowX.AI Designer → Platform Status.
Upgrade overview
FlowX.AI 6.0 is a feature release. On SaaS, FlowX applies every change on this page. Self-hosted deployments usually follow LTS releases and receive these changes with the next LTS release family. If you upgrade a self-hosted installation to 6.0.0, apply the changes below in the order given in Migrating from v5.9.x to v6.0.0.Before you upgrade a self-hosted installation
- Set every AI service group flag explicitly. See AI service groups in the Helm chart.
- Deploy designer-user, the new
coreservice. harness and ui-driver (devtime) and dynamic-classifier (runtime) deploy only if you turn their group on. See New services. - Create the new database users and secrets for the services you deploy. Every installation needs the four new keys in
keycloak-flowx-realm-clients. See New databases and secrets. - Give process-engine read access to the integration-designer bucket. See New databases and secrets.
- If you turn on dynamic-classifier, set its bucket name and provide its 20Gi volume. See Update your values.
- If you route with Kubernetes Ingress, move callers off the admin Ingress routes that the chart no longer creates. See New databases and secrets.
- Create the new Kafka topics. If your cluster uses ACLs, also give the new clients of three existing topics access. See New Kafka topics.
- Let audit-core reach Redis. It now uses Redis only to choose which replica runs its nightly cleanup job, not as a cache, and the chart connects it to the same Redis server as the other services. If you restrict which services can reach Redis, add audit-core. See Configuring Redis.
- Set the personal-data encryption master key on integration-designer. See Personal-data encryption master key.
- Choose how Elasticsearch replicates new indices. See Elasticsearch replica auto-expansion.
- Add the MongoDB hosts on your private network that data sources use to
FLOWX_SSRFALLOWEDHOSTS. See nosql-db-runner host restrictions. - Trust the certificates of the IMAP servers your email triggers read from. See IMAPS certificate verification.
Component versions
The versions below are the component versions that make up FlowX.AI 6.0.0, next to FlowX.AI 5.9.6, the latest 5.9.x release a self-hosted installation upgrades from. A dash marks a component that is new since 5.9.6. On SaaS, FlowX manages these versions.Core components
AI base components
AI dev-time components
AI runtime components
AI extensions
In FlowX.AI 6.0, the Observatory components run on environments operated by FlowX. They aren’t included in client deployments in this release.
Embedded components
- SpiceDB: see the 5.9.0 deployment guidelines for details.
- Qdrant (vector database): bundled in the
flowx-aiHelm chart and deployed with the AI base group (see AI service groups in the Helm chart). Its version is managed by the chart and not pinned here. See the Qdrant setup guide for deployment and configuration.
Renderers
The web SDKs ship at the designer version (11.252.0 for FlowX.AI 6.0.0). The native iOS and Android renderers version independently.
The solution-portal container app tracks the designer version.
Configuration changes
This section compares environment variables, Kafka topics, and theflowx-ai Helm chart with FlowX.AI 5.9.6. It includes the changes that reached SaaS in 5.10 to 5.13, so a self-hosted upgrade from 5.9.6 needs only this page.
New Kafka topics
Compared with 5.9.6, create these 42 topics. Create all of them, whatever your group flags: topics for services you don’t deploy stay unused. On SaaS they are created and managed automatically. AI nodes as durable jobs:
Deep research node:
Validation findings:
Structured template data model:
Runtime content import and export:
Document classification and extraction:
Event-triggered workflows:
Workflow schedules:
Bulk instance migration and move token:
Run auditing:
Designer notifications and preferences:
If your Kafka cluster uses ACLs, also give the new clients of these existing topics access:
ai.flowx.application-version.sync.out.v1: document-plugin and notification-pluginai.flowx.application-version.correction-after-app-operation.response.v1: document-plugin and notification-pluginai.flowx.application-version.resources-usages.sub-res-validation.response.v1: document-plugin
New services
Compared with 5.9.6, theflowx-ai Helm chart adds four services. On SaaS they are provisioned and managed by FlowX.
The databases, secrets, and volume are in New databases and secrets.
The harness reuses the platform’s existing Redis, so nothing new is provisioned. It uses Redis to stream a running turn to every open Designer session and to cancel a turn from any harness pod. Without Redis, turns still run, but live progress and cancellation across pods don’t work.
The harness also consumes the existing
ai.flowx.llm.config.changed.v1 topic, so a change to an organization’s AI model settings reaches it right away. Without Kafka, it picks up the change when its cache expires.
AI service groups in the Helm chart
Theflowx-ai Helm chart groups the services into four groups. Each group has its own flag, groups.<group>.enabled:
The groups are the same as in 5.9.6. The 6.0.0 chart adds designer-user to
core, harness and ui-driver to devtime, and cee-pii and dynamic-classifier to runtime. The base group is unchanged.
Set all four flags in your values. core is true in every combination:
New databases and secrets
Compared with theflowx-ai Helm chart in 5.9.6, these services need new databases. Create a database only if its service’s group is on. designer-user is in core, so every installation needs its database:
Each database has the same name as its user. Each secret holds the user’s password under the key
password.
The webhook-gateway also connects to the existing app-runtime MongoDB database with the app-runtime user.
The keycloak-flowx-realm-clients secret needs four new keys, one for each new service account:
Other chart changes:
- process-engine needs read-only access (
s3:GetObject) to the integration-designer bucket, to download the Kafka payloads that integration-designer stores there. The chart reads the access key ID fromglobal.flowx.s3.credentials.processEngine.accessKey(defaultflowx-process-engine-accesskey) and the secret access key from theflowx-process-engine-s3secret, keySecretAccessKey. - dynamic-classifier stores its data on a 20Gi
ReadWriteOncepersistent volume claim,dynamic-classifier-data, and uses theRecreatedeployment strategy, so an upgrade stops the old pod before the new one starts. - dynamic-classifier reads an OpenAI API key for its LLM judge from the
dynamic-classifier-openaisecret, keyOPENAI_API_KEY. The secret is optional: without it, the pod still starts and classification keeps working. - The chart no longer creates an admin Ingress for doc-converter, data-privacy, speech-to-text, web-crawler, and dynamic-classifier. ai-gateway keeps its own.
New required secret: personal-data encryption master key
The Mask Personal Data and Restore Personal Data nodes encrypt the placeholder map they keep during a workflow run. The key material comes from a new integration-designer variable with no default: without it, the mask and restore nodes fail rather than encrypt personal data with a published value. On SaaS this secret is provisioned and managed by FlowX. New integration-designer variables:New object storage bucket: runtime content
Runtime content export and import stores its files in a new shared bucket,runtime-content-bucket. Application Manager creates the bucket on first use, Integration Designer writes export files and reads imports there, and the embedder reads and writes Knowledge Base vectors at the paths each request names. Grant all three services access to the bucket.
Browsers upload import files and download export files directly, through presigned URLs that Application Manager signs. By default it signs them with the internal storage endpoint, which only pods can reach, so set APPLICATION_FILESTORAGE_S3_PRESIGNSERVERURL to a storage endpoint that browsers can reach.
New variables on application-manager and integration-designer. Set the bucket on both, and the presign settings on application-manager only:
For the full list of buckets, see Data architecture.
Changed default: Elasticsearch replica auto-expansion
New process instance indices and new audit log indices now use the Elasticsearchauto_expand_replicas setting, with a default of 0-1. Elasticsearch keeps 0 replicas when the cluster has one data node and 1 replica when it has two or more, and adjusts the count as nodes join or leave. On SaaS, Elasticsearch is managed by FlowX.
New process-engine and audit-core variable:
- Overrides the fixed replica count: while auto-expansion is on,
FLOWX_ELASTICSEARCH_INDEXSETTINGS_REPLICAShas no effect on new indices. - New indices only: the setting applies to the
process_instanceindex and index template and to the audit logs data stream template. Existing indices keep their current settings. - Disk usage: on clusters with two or more data nodes, the replica roughly doubles the disk space these indices use.
Changed behavior: nosql-db-runner host restrictions cover MongoDB
The host restriction that relational data sources already had now applies to external MongoDB connections too. The nosql-db-runner checks every address a MongoDB host resolves to, including replica-set members and SRV targets, and rejects the connection if any of them is private. For FlowX Database and Unmanaged MongoDB data sources alike, it also rejectsUPDATE, UPDATE_MANY, DELETE, and DELETE_MANY operations with a missing or empty filter. To match every document, pass {"_id": {"$exists": true}}.
nosql-db-runner variables (both already existed for relational data sources):
Changed behavior: name placeholders in data-privacy
When data-privacy masks a person’s name, the placeholder now carries the gender, for example<NAME_F_1> or <NAME_M_1>. When the name doesn’t show the gender, the placeholder stays <NAME_1>. Prompts or scripts that match the old placeholder form need updating.
Changed behavior: Workspace operations editor role replaced
The upgrade replaces the Workspace operations editor role with the Workspace corrective actions editor role. Users and groups that had the old role get the new one automatically and keep access to process instance migration and move token. No action is needed.New environment variables: audit index retention
The Audit service can now delete old audit data and roll over by shard size. Both settings are off by default, so the Audit service doesn’t delete data unless you set a retention period. New audit-core variables:Changed default: IMAPS certificate verification in email-gateway
The email-gateway now verifies the certificate of every IMAPS server it connects to. A server must present a certificate signed by a trusted authority and matching its host name. New email-gateway variable:Changed default: AI node retry
AI nodes without retry settings of their own now retry when a dependency is temporarily unavailable, such as an LLM provider outage, a rate limit, or a lost connection. Permanent errors aren’t retried, and deep research nodes never retry by default. A node’s own retry settings replace the default. New integration-designer variables:Upgrading from an older patch release
Upgrading from 5.9.5 or earlier? Apply the 5.9.6 deployment guidelines as well. Two 5.9.6 changes affect the upgrade:- Build redeployment retries: the queue of build deployments waiting for a retry moved from Redis to the application-manager PostgreSQL database, with a new topic,
ai.flowx.build.deployment-failed.v1. Deployments waiting for a retry in Redis when you upgrade aren’t moved to the new queue. Re-publish those builds after the upgrade. - advancing-controller management port:
MANAGEMENT_SECURITY_ENABLEDis no longer read. Actuator endpoints on the management port, including Prometheus, answer without authentication.
For previous FlowX.AI releases, please refer to the Release Notes overview.

