[Feature] KFP pipeline artifacts remain on in-cluster SeaweedFS while syncing to external AWS S3 or Azure Blob in the background - #311
Open
shay79il wants to merge 12 commits into
Open
Conversation
shay79il
changed the title
GiladShapira94
requested changes
Aug 20, 2026
…dation for SeaweedFS remote configuration
…d, detailing local and remote storage options with configuration examples
… configuration; enhance validation checks for Azure provider
…etails; refactor static credentials usage in templates
… disable remote storage
…guration details; add nonempty option in values.yaml for handling existing pipeline data during upgrades
…structions; update values.yaml for Azure storage configuration; enhance validation checks for Azure provider in templates
…rom values files and templates; update README.md to clarify SeaweedFS usage for pipeline artifacts.
…nfiguration; enhance validation checks for SeaweedFS remote provider in templates.
… by breaking down the command into multiple lines for better readability and maintainability.
GiladShapira94
requested changes
Sep 1, 2026
… configuration; enhance Azure Blob example and clarify usage in templates.
| pipelines: | ||
| images: | ||
| mysql: | ||
| tag: "8.4" |
Collaborator
There was a problem hiding this comment.
Why did you add the image tag to the example?
Collaborator
Author
There was a problem hiding this comment.
the testing fails otherwise
you want me to remove it?
| endpoint: s3.<REGION>.amazonaws.com | ||
| region: <REGION> # e.g. us-east-2 | ||
| forcePathStyle: false | ||
| mount: |
Collaborator
There was a problem hiding this comment.
These are the default values from the values, here you should add only the one the needs to be change
Collaborator
Author
There was a problem hiding this comment.
in values.yaml the default is using local
you want me to put them in values.yaml with local?
| @@ -0,0 +1,20 @@ | |||
| storage: | |||
| mode: local | |||
Collaborator
There was a problem hiding this comment.
This part is point MLRun to store artifacts files in Azure/S3 if you want to keep it need to change the mode to the relevant provider
| {{- include "mlrun-ce.pipelines.labels" . | nindent 4 }} | ||
| type: Opaque | ||
| stringData: | ||
| accesskey: {{ include "mlrun-ce.seaweedfs.s3.accessKey" . | quote }} |
Collaborator
There was a problem hiding this comment.
From the helper, it looks like it is using MLRun configuration. Can you please verify that you can point MLRun to local but pipelines to remote
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
📝 Description
Adds SeaweedFS remote gateway support so KFP pipeline artifacts remain on in-cluster SeaweedFS (
pipelines.storage.mode: local) while syncing to external AWS S3 or Azure Blob in the background.This replaces the previous model where KFP talked directly to external object storage via
pipelines.storage.mode: s3/azure-blob. MLRun and Jupyter storage (storage.mode) is unchanged and independent.🛠️ Changes Made
seaweedfs.remote.enabled):seaweedfs-remote-secret.yaml— cloud credentials (S3 keys fromstorage.s3.*, Azure fromstorage.azure.*)seaweedfs-remote-config-job.yaml— Helm hook Job that runsremote.configure+remote.mounton the filerseaweedfs-remote-gateway-deployment.yaml—filer.remote.gatewaysyncs local bucket → remote backendexamples/seaweedfs-remote-s3-overlay.yaml,examples/seaweedfs-remote-azure-overlay.yamlstorage-validation.yaml):pipelines.storage.mode: local(rejects direct external KFP modes)seaweedfs.remote.*(provider, bucket, remote name charset, S3 endpoint, credentials)_helpers.tpl):mlrun-ce.pipelines.storage.modehardcoded tolocalprovidersblock fromkfp-launcher.yamlOBJECTSTORECONFIG_REGIONfromml-pipeline.yamlmlpipeline-seaweedfs-artifact.yaml→mlpipeline-artifact-secret.yamlmysql.yaml): auth plugin flag selected by image tag —--default-authentication-pluginfor 8.0.x,--mysql-native-password=ONfor 8.4+deployment.yaml+values.yaml): addedstartupProbe(/api); readiness probe path changed from/labto/apifor slow cold startsseaweedfs.remoteblock (defaultenabled: false); added tovalues.yamland all three install-mode values files0.12.0-rc.11✅ Checklist
charts/mlrun-ce/Chart.yaml.🧪 Testing
Please see