Releases: SeldonIO/seldon-core
Releases · SeldonIO/seldon-core
v2.1.0
Changelog
All notable changes to this project will be documented in this file. Dates are displayed in UTC.
Generated by auto-changelog.
v2.1.0
17 January 2023
- updated notebook markdown
#4593 - update income example and yamls
#4592 - WIP: v2 licenses update
#4588 - WIP: update copyright
#4589 - Allow ServerConfig and Server labels and annotations to be passed to StatefulSet
#4543 - Bump MLServer to 1.2.3
#4590 - Add SASL_SSL for Kafka Auth
#4536 - Add bytes rawcontents correct handling and ensure headers are passed
#4586 - AnchorText example for movies sentiment
#4582 - Fix model delete hang when FailedSchedule
#4583 - use helm to install Core v2 via Ansible
#4574 - Fix SSL modelgateway kafka admin client bug
#4584 - Update triton image to 22.11 everywhere
#4585 - introduce Snyk security scans for V2 images
#4575 - Add CLI config command to manage configs
#4538 - Bump MLServer's message size
#4571 - v2: Ensure grpc outgoing context has metadata
#4568 - Change dataflow application id to unique
#4569 - v2: update triton, grpc payload size, cli raw output processing
#4560 - Retry ModelFailed models
#4557 - bugfix: reference pipeline name after handling error
#4548 - Fix grpc connection retry to control and data plane services
#4551 - Pipeline inputs
#4499 - add10 model settings
#4546 - Update markdown transforming sed scripts
#4500 - remove need for ONESHELL in go lint
#4535 - Update local-over-commit-test.ipynb
#4534 - Add an example k8s server pvc example
#4529 - change stream id to use pipeline name
#4512 - Add data-flow 3rd party license generation
#4526 - Fix go license tool install
#4524 - Update tls licenses and add summary
#4523 - Seldon Core V2 Go 3rd party licenses
#4520 - Link how to install docker compose v2 from github releases
#4516 - Fix possible SIGSEV after producer close in modelgateway
#4515 - Ensure x-request-id header matches kafka key
#4511 - Upgrading docker compose CLI command
#4498 - Disable autoscaling tracking on agent for models that are fixed
#4501 - add a notebook test for changing model replicas
#4504 - Bump MLServer version to 1.2.1
#4503 - [SCv2] Improve inference docs re routing and headers
#4481 - Enable GPU for docker-compose deployment
#4496 - Return meaningful error on empty bucket sync for model uri
#4487 - v2: Autoscaling docs - small updates
#4491 - improve landing README
#4493 - Temporary fix for envtest for operator
#4492 - update readme
#4490 - Fix model load order and delete rclone folders
#4477 - Update API protocs with latest protoc
#4476 - do not include V1 changes in V2 changelog
#4473 - Generating changelog for v2.1.0
428cbb8 - Setting version for helm charts
2502ad6 - Setting version for yaml manifests
21e2970
v1.15.0
All notable changes to this project will be documented in this file. Dates are displayed in UTC.
Generated by auto-changelog.
v1.15.0
5 December 2022
- Images update
#4463 - Bump MLServer version to 1.2.0
#4448 - add tar to wrapper image as it is missing after move to ubi-minimal
#4458 - 4041 upgrading jaeger
#4406 - Bump MLServer version to 1.2.0
#4448 - add tar to wrapper image as it is missing after move to ubi-minimal
#4458 - Removing dependabot bot
#4456 - factored out _make_rest_metrics_server
#4446 - 4041 upgrading jaeger
#4406 - Factor out _make_run_grpc_server
#4351 - update kind in ansible setup
#4440 - Add missing related images for v2 protocol
#4437 - Improve Kafka config handling in executor
#4435 - Set allowPrivilegeEscalation to false as default value for the service orchestrator
#4427 - Allow urllib3 >= 1.26.5
#4394 - push of MLServer images for Red Hat certification
#4417 - Enable SSL_SASL for executor request logging
#4416 - scan -sc images only for 1.15.0
#4419 - add mlserver-sc and mlserver-sc-slim to security scans
#4418 - Update libraries to solve CVEs reported for 1.15.0 build
#4405 - Openshift move to quay
#4392 - Fix Conda download version
#4393 - Fix vim CVE in Alibi images
#4343 - Reverted Conda base image to 4.13.0 until patched in 4.14.0
#4390 - Ambassador v2 support
#4290 - factored out make_rest_server_debug/prod
#4268 - update k8s versions
#4350 - Use timer instead of time.After to prevent memory leaks in logger
#4338 - Fix resource lock type
#4342 - Embedding intro video in docs
#4337 - update conda in base image and use ubi9
#4329 - Add reference for V2 Inference Protocol
#4325 - Add progress deadline support for SDeps
#4235 - Change service key to allow container services to always match correctly
#4043 - added comments about routing in predictChildren
#4267 - Add UTF-8 support for Flask jsonify
#4271 - Ensure request is chained before payload is logged
#4301 - Adding licenses updates as part of the post-release process
#4289 - add upgrading note on Flask 2.x upgrade
#4288 - move to Flask 2.x
#4286 - 1.14.1 to master
#4287 - Added fix for clashing zombie webhook
#4265 - Make verbosity configurable and not leak sensitive values
#4249 - Added fix for webhook issues on 1.12.0
#4256 - Update stalebot.yml
#4250 - Adding prepackaged server separate pod instructions
#4238 - doc: add util comment && identation
#4242 - Adding stalebot for issues and PRs with defaults
#4232 - Fixed trailing dash created from helm split resources
#4230 - Fix Typo in Readme.md
#4228 - enh: Add support to configure PrepackedTriton with no storage initialiser
#4216 - Added fix for removed guard on webhook
#4218 - fixes foldering of the gpt2 minio notebook
#4197 - Allow leader election controls for manager
#4211 - factored out parse_args
#4213 - upgrade pip, conda and setuptools in s2i image
#4210 - Fix logging args.grpc_workers
#4212 - renamed server_[123]func to server[rest|grpc|custom]_func
#4214 - typo fix in logging bind_address of gRPC server
#4200 - fix metadata
#4207 - typo fix in logging number of gRPC threads used
#4194 - typo fix in logging number of gRPC workers
#4195 - fix link to minio example in triton page
#4196 - Added 1.15.0-dev tag
#4174 - add missing yaml styling for snippets
#4170 - update rest_predict_seldon hardcoded version in route
#4161 - update licenses for 1.15.0 release
449510d - Release 1.15.0
3502d35 - new AWS installation guide and updated nav to reflect it
127ce8d
v2.0.0
Changelog
All notable changes to this project will be documented in this file. Dates are displayed in UTC.
Generated by auto-changelog.
v2.0.0
2 December 2022
- do not include V1 changes in V2 changelog (#4473)
#4474 - fix image build
#4470 - Update docs to reference new SCv2 location
#667 - Rename Go modules for repo migration
#666 - Improve logging for config and log levels
#453 - Remove versions dropdown and align top bar with SCv1 docs
#665 - Fix nil pointer exception on scheduler restart
#664 - Add ReadTheDocs config file
#663 - remove unnecessary event on unload
#662 - Fix http calls mirror pipelines not working
#660 - Fix experiment stop
#661 - Agent grpc server max connections
#655 - adjust workflows for migration
#654 - Fix Server replica Helm templating
#648 - Add new constant rate scenario
#652 - revert pipeline create changes
#651 - use consistent name for consumer groups
#649 - k6 docker file fixes
#637 - Change to use mlserver 1.2.0
#647 - Add ability to wait on 3 termination for drain
#644 - Add Helm parameterisation for server replicas
#643 - Update trailer check to one if block
#642 - Add huggingface capability to Helm charts
#641 - only add trailers headers if not nil
#640 - set parallel workers to zero for explainers and update notebooks
#632 - Use kafka 3.3.1 by default
#631 - Prevent terminationGracePeriodSeconds being treated as string in helm charts
#628 - Run
misspell -w .on docs#626 - Fix typos in pipeline docs
#625 - fix type of terminationGracePeriodSeconds in helm charts
#623 - Pipeline Readiness
#547 - HPA server autoscale
#590 - MLServer update to 1.2.0.rc5
#617 - set chart version also for seldon-core-v2-certs
#620 - Add validation for empty pipeline steps
#607 - Adding
UnloadEnvoyRequestedmodel replica state#616 - Add securityContext parameterisation to Kubernetes manifests for OpenShift compatibility
#606 - Run all notebooks and updates to pipeline validation
#597 - Model gateway logs
#613 - Fix k8s version updates
#612 - Respect MLServer content type in pipeline gateway http server responses
#600 - Add omitempty tag to parallel_workers field
#603 - Add resource parameterisation for all components in Helm chart
#596 - remove duplicated seldon-v2-crds file
#594 - Ensure kafka consumer reconnects happen by not ending consumers
#595 - Add pipeline state to k8s resource status
#591 - Fix pipeline http calls to use headers and reintroduce prom metrics removed
#580 - Revert to vanilla notebook to markdown conversion
#585 - Update kind and k8s versions in Ansible setup
#586 - Return model name in pipeline errors
#583 - Add Apache 2 Licence to code files
#584 - Override config file with CLI args for Seldon CLI
#579 - add rolling update md
#582 - Updates for Model rolling updates
#566 - Quickfix/cli pipeline error
#572 - Quickfix/agent scheduler restart
#568 - Issue 561 pipeline err
#569 - Remove redundant pipeline gateway and agent Prometheus metrics
#554 - Ensure model gateway keeps recalling scheduler and does not restart
#563 - Docs fix for readthedocs static images
#565 - Fix static image uris
#564 - Quickfix/fixstop notready
#559 - Updating styles + new images
#548 - Agent startup bug fix
#558 - only output warning in case of error
#556 - Add missing Ansible config vars + fix incorrect docs on metric names
#553 - Remove namespace from Prometheus metric names
#550 - Issue 518 agent stop cmd
#523 - small fixes for huggingface demo
#549 - Fix initial routes for Pipelines and delete of Pipelines from Envoy
#543 - cert download script and docs
#544 - Add LICENSE and script to add copyright to Go files
#542 - fix space in go file name
#541 - Speech to Sentiment Example Updates
#540 - Quickfix/add paper reference
#525 - Huggingface Speech to Sentiment Example
#519 - Bump MLServer version
#538 - small fixes for mlserver, docs and envoy yaml
#537 - [CLI] Support authority headers for control-plane subcommands
#531 - Add missing Envoy patch file for Kustomize generation of Helm charts
#536 - [CLI] Use positional args for resource name in CLI server-status subcommand
#533 - [CLI] Use cURL-style request metadata logging for gRPC
#528 - Add gRPC service name prefix in components Helm chart
#530 - Add initial drift and outlier docs
#529 - [CLI] Support authority headers for inference requests
#526 - Use cURL-style request metadata logging in Seldon CLI
#524 - Add Helm parameterisation for scheduler and Envoy service types
#520 - Add a note about resetting model autoscaling
#517 - Update autoscaling docs
#514 - Issue 507 model autoscaling docs
#513 - Add huggingface as runtime and example
#511 - Install doc updates and developer doc additions
#510 - Add server and certs Helm charts and raw manifests to published assets
#508 - ...
v1.14.1
5591c42
This commit was signed with the committer’s verified signature.
The key has expired.
axsaucedo
Alejandro Saucedo
v1.14.1
17 August 2022
- Added fix for clashing zombie webhook
#4265 - fix workflow on red hat image scan process
#4259 - Make verbosity configurable and not leak sensitive values
#4249 - Make verbosity configurable and not leak sensitive values
#4249 - Added fix for webhook issues on 1.12.0
#4256 - Update stalebot.yml
#4250 - Adding prepackaged server separate pod instructions
#4238 - doc: add util comment && identation
#4242 - Adding stalebot for issues and PRs with defaults
#4232 - Fixed trailing dash created from helm split resources
#4230 - Fix Typo in Readme.md
#4228 - enh: Add support to configure PrepackedTriton with no storage initialiser
#4216 - Added fix for removed guard on webhook
#4218 - fixes foldering of the gpt2 minio notebook
#4197 - Allow leader election controls for manager
#4211 - factored out parse_args
#4213 - upgrade pip, conda and setuptools in s2i image
#4210 - Fix logging args.grpc_workers
#4212 - renamed server_[123]func to server[rest|grpc|custom]_func
#4214 - typo fix in logging bind_address of gRPC server
#4200 - fix metadata
#4207 - typo fix in logging number of gRPC threads used
#4194 - typo fix in logging number of gRPC workers
#4195 - fix link to minio example in triton page
#4196 - Added 1.15.0-dev tag
#4174 - add missing yaml styling for snippets
#4170 - update rest_predict_seldon hardcoded version in route
#4161 - certified operator: create bundle (step 1)
602b4c2 - update 1.14.0 community bundle for OpenShift
05ec72b - release 1.14.1
eefa9e4
v1.14.0
8307fe3
This commit was signed with the committer’s verified signature.
The key has expired.
axsaucedo
Alejandro Saucedo
v1.14.0
16 June 2022
- Fixed operator redhat image
#4157 - fix broken mlflow model build
#4155 - Bump MLServer version to 1.1.0
#4148 - Upgrade to k8s 0.23 APIs, remove v1beta1 as default, upgrade KEDA
#4136 - Create graph-modes.md
#4144 - Fix typo in error message for Anchor tabular
#4145 - fix transport missing in executor
#4107 - fix alibi tests
#4142 - Broken docs test fix removing reference to Tree
#4141 - Fix. Ensemble model. Previouse not saved data in jagear. Working with Jagear and Istio
#4139 - Add optional manual commit to seldon kafka server
#4117 - update kind
#4135 - use alternative multiprocessing library if USE_MULTIPROCESS_PACKAGE i…
#4114 - upgrade alibi explain to 0.7.0
#4112 - Update cache folder and bump MLServer image
#4094 - Adding protocol info to executor payload logging worker
#4077 - Don't hardcode UID for Triton containers
#4099 - Sorted metric tags to avoid duplicate prom data with gRPC requests
#4006 - respect envSecretRefName coming from helm values
#4089 - minor type fix
#4086 - Huggingface optimum prepackaged server
#4081 - Adding configuration for feature level drift metrics
#4079 - Fixed random seed for anchor explanation
#4078 - Re-setting numpy random seed to zero on every explain request
#4076 - Pass through model name env var for MLServer
#4069 - Update seldon-deployment.rst
#4075 - Adding tests to explicitly state expected behaviour of v2 protocol chaining in REST
#4061 - add prometheus operator docs
#4038 - change versions we test upgrade of operator from
#4066 - lock jager operator helm chart to fix integration tests
#4064 - Protocol specific ready checkers
#4028 - fix(executer): Forward parameters while chaining models via kfserve grpc
#4054 - Outlier example poetry
#4055 - fix removal of request logger to fix CI
#4044 - Enabling optional grpc server on python level only
#4027 - Removed request logger from github security workflows
#4039 - Python request logger example component deprication and removal
#4016 - Extended GPT2 MLServer Pipeline Example to include post-processor
#4035 - Updated CPP example to use latest 3.8 base image
#4026 - allow priorityClassName for manager
#4030 - Updating model inputs for new schema
#4032 - Updating typings for prediction API documentation
#4025 - Fixed Flask breaking version by werkzeug dependency limit to 2.1
#4018 - Update ab_test_2pods.json
#4020 - Decompress prediction events before logging to kafka topics
#4005 - Updated poetry environment and lockfile for Alibi Detect 0.9.0
#4001 - update sklearn iris example
#3995 - Release v1.13.1 for OpenShift
#3987 - Bumping rclone image version to 1.57.0
#3990 - fix example yaml file error in README.md
#3994 - added tag for s2i python image
#3992 - Fixing nbqa linting for latest notebook
#3991 - support traffic settings for shadow deployment with istio
#3780 - Issue #3968: Allow hostNetwork=true for seldon operator
#3971 - Updating explainer docs into 0.6.4
#3976 - Add TLS to Kafka Consumer and also add Kafka + KEDA + TLS example
#3977 - Merging 1.13.1 changelog and update to 1.14.0-dev images
#3962 - Bumping rclone image version to 1.57.0 (#3990)
#3973 - Updating changelog to 1.13.1
adf6c54 - Release v1.14.0
158950f
v1.13.1
v1.13.1
21 February 2022
- Updated base golang images to 1.17.7
#3951 - cast float/int 64 to 32 in alibi-detect-server
#3958 - Update security policy to outline current security scans
#3959 - Addresing security vulnerabilities for 1.13.1 release
#3949 - Updating broken link in documentation
#3950 - Fixed Flask dependency by pinning markupsafe and itsdangerous
#3948 - Fixing failing docs CI tests
#3915 - Update images to 1.14.0-dev for next semver release
#3939 - Adding chmod to dockefile example
#3937 - Update Adopters.md
#3934 - Update kfserving-storage-initializer.md
#3831 - Release v1.13.1
c696e99 - Release v1.13.1 security images
d90766d - Release v1.13.1 changelog
83f2f63
v1.13.0
v1.13.0
- Add test for tensorflow prepackaged Seldon protocol with resource requests specified
#3928 - Bump MLServer to 1.0.0
#3927 - Skip request logging if skip header is present
#3925 - upgrade alibi explain to 0.6.4
#3885 - Allow v2 as protocol name
#3906 - Bump MLServer image to 1.0.0.rc2
#3916 - Update gcp.rst
#3921 - Add model_name when chaining requests
#3805 - Fixing Alibi Detect Server response cloud event data is json marshalled string
#3907 - bump alibi-detect to 0.8.1 in adserver
#3871 - Updating inference logic to add node level request-response logging
#3874 - Pass down ports info to MLServer
#3898 - Update autoscale docs
#3905 - Updating helm docs for 1.13.0-dev
#3879 - Updated cert-manager API version
#3888 - Fix seldon manager configmap for alibiexplainer version
#3897 - Adding test skip until fixed via 3857
#3894 - Redhat 1.12.0
#3878 - Add support to use PEM string for SSL
#3868 - Changes ndarry to ndarray
#3892 - 3804 Removal of Depricated Java Engine Resources
#3845 - Updating security tests to run on 1.13.0-dev images
#3875 - Upgrade confluent-kafka-go to v1.8.2
#3870 - Bump upper constrains of MLflow server dependencies
#3863 - Add events to namespaced roles
#3855 - Seldon add ssl
#3813 - Add agrski as approver
#3865 - Operator sets seldondeployment to failed when deployment not progressing
#3851 - Read OIDC resource parameter
#3844 - Update mlflow.md
#3843 - Update overview.md
#3842 - Remove triage label and release notes block from templates
#3835 - Bump seldon-deploy-sdk to 1.4.1.2 in request logger
#3838 - Added missing words
#3837 - reference to kfserving storage initializer from dockerhub
#3832 - Updated license branch from master to main for hashicorp/go-version
#3829 - Update Dockerfile in python docker wrapper docs
#3822 - Fix broken link
#3820 - Exclude caBundle field when cert-manager is enabled
#3807 - Use default PID not UUID for worker ID
#3801 - Updating tag to 1.13.0-dev + adding changelog
#3799 - Add note for MacOS users
#3800 - Updated for changelog generator to use auto-changelog
0ea7e2c - Updated changelog
e6a60ae - Updated changelog with full changes!
56bcce3
v1.12.0
v1.12.0
Fixed bugs:
- Ensure docker images have the source code included for MPL based licenses #3787
- Stateful Model Feedback Metrics Server uses wrong image #3784
- seldon-core-microservice: error: unrecognized arguments: test REST --service-type MODEL --persistence 0 #3775
- Tensorflow protocol can't support /v1/models/${MODEL_NAME}[/versions/${VERSION}|/labels/${LABEL}] when using istioingress gateway #3767
- seldonio/mlflowserver:1.12.0-dev now unable to server models #3766
- TEP PY SITH #3752
- OOM when stress test the Seldon model, which may be caused by the logging of request and response payloads #3726
- Unable to install seldon via Helm #3725
- Conda pip install permission denied in OpenShift #3712
- Python/Executor path mis-match #3705
- seldon-core-operator helm chart - in single namespace mode the role seldon1-manager-role does not include poddisruptionbudgets #3692
- Currently engine related values cannot be ommited from the values.yaml while deploying seldon #3691
- apiversion is deprecated in
notebooks/resources/ambassador-rbac.yaml#3677 - Error to build env: Command 'conda run -n mlflow pip install -r /microservice/requirements.txt' returned non-zero exit status 1. #3670
- Seldon Manager Config Map has an extra comma that can cause parsing error #3652
- Validate and address recent apparent flakiness in v2 server tests #3639
- Executor returns minimal information on http client failure #3625
- helm chart not compatible with k8s v1.22 #3618
- Upgrade Grafana chart in seldon-core-analytics #3558
- Mlflow getting downgraded #3454
- Deploying custom MLflow model - stuck at "Readiness probe failed" #3186
- Java Engine and Go Executor Does not Terminate Graph upon Error #2480
Security fixes:
Closed issues:
- Exposing GRPC timeouts in Python client #3779
- remove explainer_examples notebook #3777
- Submit cloud events uncompressed until consumers supports it #3749
- When is seldon v1.12 going to drop? #3735
- Add model chaining example for the V2 protocol #3732
- Fix
alibi-explain-serverto usealibiload\_explainerhelper function #3708 - Batch processor enhancemenst through raw data parameter #3702
- Allow for Annotations and Labels to be injected into helm chart templates #3699
- Allow Controller Manager to have configurable container security context #3698
- Batch processor cast integers to floats in payloads #3681
- Integrated alibi-explain mlserver runtime in Core #3675
- Add e2e example on training explainers with Poetry-locked environment #3664
- Move adserver to use Poetry #3660
- Update Openshift OperatorHub Releases for 1.11.1 #3563
- Better support Istio Mesh Internal to cluster #3485
- Enable index filed in batch_processor #3409
- Add "names" field in batch_processor.py to align with seldon_client.py #3408
- upgrade seldon-core-analytics helm charts #3403
- batch processor: add raw inputs support to predict method #2657
- Publish RedHat images via API #2085
- Seldon deployment name in case of complex graph #1801
Merged pull requests:
- fix missing rclone in adserver #3796 (RafalSkolasinski)
- Bump MLServer image to 1.0.0.rc1 #3795 (adriangonz)
- Tweak script to generate MLflow environment tarball #3794 (adriangonz)
- Addressed vulnerabilities with snyk for containers and libraries #3793 (axsaucedo)
- use lowercase seldon-puid in batch processor #3791 (RafalSkolasinski)
- batch processor: always prioritize tags from model #3789 (RafalSkolasinski)
- Added MPL Licenses to Docker images #3788 (axsaucedo)
- build adserver dockerimage using poetry-defined environment #3783 (RafalSkolasinski)
- remove explainer_examples.ipynb from notebook test #3778 (sakoush)
- Support https on batch processor #3774 (majolo)
- Docs: Removed argument from tags example #3771 (jacobmalmberg)
- Allow deployment name as prefix when default name to long #3768 (cliveseldon)
- upgrade alibi to 0.6.2 #3765 (sakoush)
- fixed iris sklearn model example #3762 (karan6190)
- Address MLServer flakiness in CI tests #3754 (adriangonz)
- executor: decompress gzip paloyads before logging them #3746 (RafalSkolasinski)
- Fixing docs-build test #3745 (edshee)
- Sherif akoush/sc 2549/alibi v1 fixes #3744 (sakoush)
- Transformers example #3743 (adriangonz)
- Chart annotations cont, manager container security context #3741 (mwm5945)
- Logger params #3738 (cliveseldon)
- fix python lint #3737 (cliveseldon)
- Fix 1.22 Webhook Create by ensuring v1 is used #3736 (cliveseldon)
- Performance fix for the logger in the executor #3734 (ivan-valkov)
- add endpoint to match python status endpoint #3731 (mwm5945)
- Mnist e2e triton #3728 (RafalSkolasinski)
- Chart annotations #3724 (mwm5945)
- Batch processor enhancemenst through raw data parameter #3718 (RafalSkolasinski)...