Repository navigation
Conversation
- Internal metadata store is a KubeDB Etcd: require the Etcd feature gate, drop the upstream etcd-io operator step. - MinIO tenant: override the unpullable default image, note one tenant per namespace. - One storage secret (distinct rootPath) per Milvus instance; distributed examples use milvus-cluster-storage-config. Document that object-storage data is not deleted with the database and how to remove a rootPath prefix. - Fix etcd PVC naming, realistic readiness time, mention <db>-etcd-auth. - README: list all catalog versions, drop Grafana claim. - concepts/milvus.md: document spec.configuration, serviceTemplates and metaStorage sub-fields.
…guides - Monitoring yamls no longer need TLS; add apply and Prometheus checks. - TLS yamls drop the undocumented monitor block; ClusterIssuer CA secret goes in the cert-manager namespace; show how to find the config secret. - Reconfigure TLS: deploy a non-TLS database first, add new-issuer.yaml and its CA, reorder the distributed flow, show serial/issuer checks. - Recommendation: create the Issuer the sample yamls reference.
- Every guide deploys the database it operates on and cleans up both milvus-standalone and milvus-cluster with WipeOut. - Reconfigure: drop restart "false" (Milvus has no dynamic reload), explain how configSecret/applyConfig/removeCustomConfig combine, verify the running pod. - Rotate auth: read auth-active-from from the new secret / spec, check the old password is rejected. - Update version: add 2.6.9 manifests, remove stale distributed note. - Horizontal scaling: yaml now requests 2/2, add scale-down yaml, document the dataNode key. - Vertical scaling: InPlace sample requests new values, show how to confirm no restart. - Volume expansion / storage migration: add source manifests (longhorn-custom, 2 streamingnodes) and missing apply commands.
- Compute: podLifeTimeThreshold 10m, resourceDiffPercentage 20 and a 1Gi memory floor to stop re-scaling after each restart; real status shape; no VPA object is created; warn that each OpsRequest restarts pods and that deleting the autoscaler deletes its running OpsRequests. - Storage: how to enable and check the storage metrics server, remove the stray compute autoscaler from the sample output, deploy steps.
There was a problem hiding this comment.
Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota.
An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing.
|
Visit the preview URL for this PR (updated for commit 69f21c8): https://kubedb-v2-hugo--pr1085-milvus-guide-test-fi-sq5f6iza.web.app (expires Wed, 04 Nov 2026 13:20:12 GMT) 🔥 via Firebase Hosting GitHub Action 🌎 Sign: 0f29ae8ae0bd54a99bf2b223b6833be47acd5943 |
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
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.
Fixes the Milvus user guide (
docs/guides/milvus) after following every page literally, standalone and distributed, on a k3d cluster with KubeDB v2026.10.12.Main fixes
Etcd, so the install needs--set global.featureGates.Etcd=true; the upstream etcd-io operator is not used and its step is removed. The MinIO tenant's defaultquay.io/minio/minioimage now returns 401, so the command overrides it. Notes that the MinIO operator allows one tenant per namespace.rootPathper Milvus: two instances on the same path collide (GC and WAL locks).milvus-standalonekeepsmilvus-storage-config,milvus-clusteruses a newmilvus-cluster-storage-config. Documents that deleting a Milvus leaves its objects behind and how to clean arootPathbefore reuse.cert-manager);spec.tlsremoved from the monitoring YAMLs andspec.monitorfrom the TLS YAMLs so files match the pages.restart: "false"means Milvus keeps running the old values (no dynamic reload); examples now restart and check the running pod.auth-active-fromis read from the right object.dataNode(camelCase); concepts page fixed.podLifeTimeThreshold: 10m,resourceDiffPercentage: 20, memory floor 1Gi (the old 1m / 10 / 256Mi values made it re-scale and restart pods every few minutes); real status shape; no VPA object; warnings about restarts and deleting the autoscaler mid-OpsRequest.spec.configuration,serviceTemplatesandmetaStorage.Every new or changed Milvus/Issuer YAML passed
kubectl apply --dry-run=server.Related product fixes (separate PRs)