[Sync] kserve/modelmesh-serving-v0.11.0-rc0 to main branch #107
Conversation
This file contains 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
Multi-platform build for `modelmesh-controller` and `modelmesh-controller-develop`; enable minikube deployments on Mac M1 laptops with ARM chip (opendatahub-io#162, opendatahub-io#231) - Remove `amd64` arch-specific `nodeAffinity` for modelmesh controller (opendatahub-io#162, opendatahub-io#231) - Build for platforms `linux/amd64`, `linux/arm64`, `linux/ppc64le`, `linux/s390x` - Use `ubi8/go-toolset:1.18` for developer image instead of `ubi8/ubi-minimal:8.7` - Build on `pull_request` as well as on `push` to catch build breaks during PR review - Push developer images to DockerHub on PR merge, use local registry for PR builds - Update deprecated `checkout` action - Fix infinite docker inside docker error when running `make run fmt` inside dev container - Update `build_devimage.sh` script to only pull developer image if not already present - Update `develop.sh` script to use the `.develop_image_name` Resolves opendatahub-io#162 Resolves opendatahub-io#231 --------- Signed-off-by: ddelange <14880945+ddelange@users.noreply.github.com> Signed-off-by: Christian Kadner <ckadner@us.ibm.com> Co-authored-by: Christian Kadner <ckadner@us.ibm.com>
Signed-off-by: Christian Kadner <ckadner@us.ibm.com>
Remove the `--enable-self-signed-ca` option from the quickstart guide and webhook information. The quickstart instructions explicitly install the latest release `v0.10` which does not include the recently added `--enable-self-signed-ca`. Following the document as-is results in an `Unknown option` error. Resolves kserve#370 Signed-off-by: Rafael Vasquez <raf.vasquez@ibm.com>
Fix "no space left on device" error for FVTs on GitHub action runners by deleting libraries that are not needed for dotnet, Android, PowerShell, Swift which increases available disk space by 25 GB plus an additional reclaimed space of 2.6 GB after pruning docker images. Resolves kserve#367 Signed-off-by: Christian Kadner <ckadner@us.ibm.com>
Add `arm64` to list of supported architectures for Node affinity to enable Quickstart deployments on MacBooks with M1 chips. Currently only `amd64` and `arm64` are supported by the ModelMesh runtime adapter. Signed-off-by: Christian Kadner <ckadner@us.ibm.com>
Update TorchServe image version in the default serving runtime from 0.6.0 to 0.7.1 Closes kserve#360 Signed-off-by: Rafael Vasquez <raf.vasquez@ibm.com>
Update MLServer image for the default serving runtime from 0.5.2 to 1.3.2 Depends on kserve/modelmesh-runtime-adapter#45 Closes kserve#357 Signed-off-by: Rafael Vasquez <raf.vasquez@ibm.com>
#### Motivation To port a previously-built Grafana dashboard developed for displaying various prometheus metrics for ModelMesh and build upon it to include metrics at the deployment/runtime-view #### Modifications - Added `config/dashboard/` directory to host grafana dashboard JSON - Added `servicemonitor.yaml` to `config/prometheus/` - Added and modified `ModelMeshMetricsDashboard.json` to work out-of-the-box - Added queries for deployment views to filter by serving runtime - Created sections for global-view and deployment-view visualizations - Created drop-down variables for filtering between one or many pre-existing runtimes with option to filter by custom serving runtime - Updated `monitoring.md` doc to include guide for setting up Prometheus/Grafana for monitoring; mainly using instructions from [modelmesh-performance](https://github.com/kserve/modelmesh-performance/tree/main/docs/monitoring) #### Result - A functional, ready-to-use dashboard JSON - General documentation outlining out to set up monitoring and use the JSON - Closes kserve#335 Signed-off-by: Rafael Vasquez <raf.vasquez@ibm.com>
Update TritonServer image version from 21.06.1 to 23.04 Closes kserve#358 Signed-off-by: Rafael Vasquez <raf.vasquez@ibm.com>
Update KServe and dependencies in preparation for v0.11.0 release: - Update go.mod - Update CRDs under config/crd/bases - Remove outdated module replacements - Update Dockerfile.develop: - Use Go 1.19 - Fix broken controller-gen install - Add safe.directory work-around for the git 'dubious ownership' error during the GHA `lint` workflow - Temporarily disable lint deprecation check "SA1019" - Update Copyright header formatting in *.go sources to not be mistaken as package documentation - Update mock client GET function signature in grpc_resolver_test.go Signed-off-by: Rafael Vasquez <raf.vasquez@ibm.com> Signed-off-by: Christian Kadner <ckadner@us.ibm.com> Co-authored-by: Christian Kadner <ckadner@us.ibm.com>
#### Motivation When using the `storageURI` form of an HTTP (non Azure blob) address to download a model, the `modelPath` needs to be an non-empty string. Before this change, the `storageURI: http://models.r.us/my-model.json` form would be equivalent to the following `storage` spec: ``` storage: type: http path: '' # this being empty is problematic for later processing parameters: url: http://models.r.us/path/to/my-model.json ``` The http storage type is currently the only way to have a valid storage configuration with an empty `path` (mainly because it has a "url" parameter that could include the full path). That said, I'm not sure if we should make a `path` required for the HTTP storage type. In particular, if the `url` is just `http://models.r.us/`, there is no path portion. Related: kserve/modelmesh-runtime-adapter#41 (comment) #### Modifications Set `modelPath` to the URL's Path and set the `url` parameter to not have the URL Path. #### Result With these changes, the `storageURI` example above changes to have a `path` field: ``` storage: type: http path: path/to/my-model.json parameters: url: http://models.r.us/ ``` Signed-off-by: Travis Johnson <tsjohnso@us.ibm.com> Signed-off-by: Christian Kadner <ckadner@us.ibm.com> Co-authored-by: Christian Kadner <ckadner@us.ibm.com>
Signed-off-by: Rafael Vasquez <raf.vasquez@ibm.com>
Signed-off-by: jooho <jlee@redhat.com>
Signed-off-by: jooho <jlee@redhat.com>
Signed-off-by: jooho <jlee@redhat.com>
Signed-off-by: jooho <jlee@redhat.com>
fe5db8e to
a16bbe3
Compare
&& Update runtime/fvt manifests Signed-off-by: jooho <jlee@redhat.com>
|
/retest |
Signed-off-by: jooho <jlee@redhat.com>
Signed-off-by: Jooho Lee <jlee@redhat.com>
|
@Jooho: The following test failed, say
Full PR test history. Your PR dashboard. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here. |
danielezonca
approved these changes
Jun 9, 2023
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: danielezonca, Jooho The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
|
/lgtm |
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.
Motivation
This is Release Sync. Detail information is here:
kserve/modelmeshrelease issue: release: Update image tags for v0.11.0-rc0 kserve/modelmesh-serving#390kserve/modelmesh-servingrelease version : v0.11.0-rc0odh/modelmesh-servingbranch: mainTest
Result
PR checklist
Checklist items below are applicable for development targeted to both fast and stable branches/tags
Checklist items below are applicable for development targeted to both fast and stable branches/tags