* fix(sdk): compile ParallelFor in a deterministic manner
During compilataion ParallelFor components end up with randomized names,
which makes it very inconvenient to compare two versions of a pipeline.
This commit fixes this issue.
* fix(sdk): fix new parallel-for test cases
* add test for keyword-only arguments in pipeline func
* fix: kwargs-only argument for pipeline func
* test: kwargs generate same yaml as args
* remove whole metadata
* assert -> self.assertEqual
* programmatic example --> fixed example
* same name for both
Co-authored-by: Alexey Volkov <alexey.volkov@ark-kun.com>
* SDK - Components - Fixed python components that use \n
The escape sequence was being replaced by the `echo` command.
Apparently, unlike in the `bash` shell, the `echo` command of the `sh` shell expands the escape sequences by default and does not support an option to turn it off. (For some reason the -n option works properly even though it should not).
Fixes https://github.com/kubeflow/pipelines/issues/4939
* Fixed the test data
* Fixed the deprecated container component builder
* Fixed the new compiler test case
* Added test
The component specification has always supported component annotations, but there was no way to specify them for the components generated from python. This PR fixes that.
* add placeholder to spec
* add output_directory to pipeline
* respect uri placeholder in file outputs
* wip: add data passing rewriting logic to respect the uri semantics
* merge input_uri and paths when instantiating ContainerOp
* fix
* fix workflow rewriting
* Add topology rewriting
* add a test case, and various fixes
* make the test case more complex
* Fix the case when working with OpsGroup
* Fix test case
* fix resolving test
* fix redundant cmd lines
* fix redundant cmd lines
* resolve comments
* fix file outputs
* resolve comments
* copy file outputs instead of modifying inplace.
When calling the delete() method of a ResourceOp we need to ensure we do
not wait for its deletion.
The reason for this is described in [1]: If a pipeline creates a
resource which is being consumed by its steps (e.g., a PVC), the step
deleting the resource will hang waiting for the Kubernetes resource
deletion which, in turn, is waiting for the other steps to get deleted.
As a result, the pipeline never finishes.
This commit allows specifying flags for the ResourceOp kubectl commands
and defaults to the '--wait=false' flag for the deletion.
Specifying flags for a ResourceTemplate is not supported in Argo v2.7
that we currently deploy. But they will be once we upgrade to v2.11+
[2]. This does not affect the delete() method because we don't rely on
Argo's ResourceTemplate for it.
[1] https://github.com/kubeflow/pipelines/issues/4506
[2] https://github.com/kubeflow/pipelines/issues/4553
Signed-off-by: Ilias Katsakioris <elikatsis@arrikto.com>
* allow calling ContainerOp.after with multiple ops
* m: simplify signature of OpsGroup.after
as it's not public anyway, chances of calling it with by-keyword
argument are small compared to calling it with unpacking an arbitrary
list, which could be empty (previous signature would fail for empty list)
Currently were running the python code inline using `python -c <code>`.
This has two issues:
1) Python does not show source code line in exception stack traces
2) inspect.getsource does not work. This method is used in PyTorch JIT for example.
We solve these issues by writing the code into a file before executing it.
The disadvantage of the new approach is that it adds complexity, a filesystem write operation and also requires the `sh` executable to be present (we could replace it with python-based program if needed).
* feat(sdk): add ability to set retry policy
This fixes the second part of the issue described in #4333
The first part was addressed in #4392
* feat(sdk): validate retry policy name
* feat(sdk): simplify retry policy interface
* Fix bug where we missed injecting importer node
* moved files
* address review comments
* Add InputUriPlaceholder and OutputUriPlaceholder
* support uri placholder in v2
* lint
* test
* Preserve a test case with inputPath and outputPath usage.
* fix ut
* fix import and setup
* cleanup v2.components
* address comments
* remove InputTextFile, InputBinaryFile, OutputTextFile, OutputBinaryFile
* address review comment
* Initial execution cache
This commit adds initial execution cache service. Including http service
and execution key generation.
* fix master
* fix go.sum
* update docs that still refer to KFP latest SDK reference
* Set current namespace in local KFP context if running from notebook
* Create "~/.config/kfp/" instead of ".config/kfp/"
At first it was assumed the `get_user_namespace` command would be executed from the home directory.
* Create local context file if it doesn't exist during set_user_namespace
* Grab path from LOCAL_KFP_CONTEXT when creating folder
Instead of harcoding the os.mkdirs path to `~/.config/kfp` it now grabs it from the LOCAL_KFP_CONTEXT. Also, removed path creation in `get_user_namespace` as that is now handled in `set_user_namespace`. Also, it now checks if the path exists rather than the local_context_file to remove the situation where it tries to create ~/.config/kfp/ because the context.json doesn't exist when the path does.
* add multi-user setting to healthz api
* Add http prefix to health api url
* move healtz api call to own function and fix multi_user boolean
* Fix HEALTH_PATH declaration
* Move check to Client __init__ and change get_kfp_healthz to avoid breaking in case of old apiserver image
* Add multi_user to frontend healthz
* Expose multi_user in frontend and add integration test
* Fix integration test
* Fix host hardcoding and error handling
* Handle empty API response, check if API up to date
* Fix response return
* remove API check due to empty response
* retry API call if first response empty
* retry getting healthz api if no response
* change health_api to https
The healthz_api has been returning empty responses which might be caused by sending an http request to an https endpoint. Although requests handles redirects, this commit is to test if this solves the issue.
* Add some debug info to healthz exception
* add url to debug and lower retries to 1
* Use api_client to get healthz data
* Debug info for API response
* Follow API redirect history
* Fix indentation
* Add healthz proto
* Try getting healthz api with new python backend
* Add installation of kfp_server_api in tests
* Fix incorrect setup location
* Replace old .get with new http backend .multi_user
* Code clean up
* Small fixes and TimeOutError for retries healthz api
* Remove changes to go dependencies
* Send empty proto request and fix exception client
* Remove unused commit_sha and tag_name
The default timeout value is of type timedelta which cannot be compared with a plain number later in the code.
Also replaced time_delta.seconds with time_delta.total_seconds() which gives out correct value.
Chen Sun
Vitalii Vokhmin
Michalina Kotwica
Jiaxiao Zheng
Alexey Volkov
Mansoor Baba Shaik
radcheb
Shu
Niklas Hansson
Yang Pan
Jiyong Jung
Ilias Katsakioris
Yuan (Bob) Gong
capri-xiyue
Kenta Onishi
Rui Fang
Rui Fang
hilcj
hilcj
DavidSpek
David Przybilla