Pryce
04/16/2023, 12:15 AMAkshat D.
04/17/2023, 8:22 AMEna Škopelja
04/17/2023, 10:06 AMStructuredDataset
inputs from the UI by just passing a link to the parquet file as input, but recently (as of 1.4.0 I think) I get an error message (screenshot attached). Is that something that is intentional i.e. you won't be providing support anymore?Eli Bixby
04/17/2023, 11:23 AMFlyteFile[Format]
syntax with `StructuredDataset`s? It looks like they are backed by a different proto, so it's not clear to me how that works.Laura Lin
04/17/2023, 8:34 PMGreg Friedland
04/17/2023, 10:36 PMRuntimeError
. According to the docs, I thought this would be considered non-recoverable and thus not retried? I’l put the stack trace in the thread.Viljem Skornik
04/18/2023, 4:00 PMKetan (kumare3)
Eduardo Matus
04/18/2023, 7:11 PMAndrew Achkar
04/18/2023, 8:36 PMKetan (kumare3)
Brian Tang
04/19/2023, 7:33 AMserviceaccount
being used by my task as trainer-service
. however, it’s not being applied to the pod. is this a bug, or am i missing something? the service account used by the pod still remains as default
. my current pod template:
@task(
requests=Resources(cpu="250m", mem="512Mi"),
pod_template=PodTemplate(
pod_spec=V1PodSpec(
service_account_name="trainer-service",
containers=[],
tolerations=...
)
Jannis
04/19/2023, 7:33 AM@task
def test(a, b, c):
return a+b+c
@workflow
def workflow_test():
kwargs = {"a": 4, "b": 3, "c":0}
test(**kwargs)
Thanks already!Serge Rubinstein
04/19/2023, 1:16 PMNo Flyte entities were detected. Aborting!
it is coming from pyflyte register
i keep getting this error on a fresh setup, but the same project keeps being processed just fine on the previous one, and i can't figure out the difference
a more general question would be: what does pyflyte consider its 'entity'?Blair Anson
04/19/2023, 1:33 PMcluster_resource_manager
templates.
My template looks like this, but I never see a pvc created when tasks are running...
- key: ac_project_efs_shared
value: |
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
name: efs-shared
namespace: {{ namespace }}
spec:
accessModes:
- ReadWriteMany
storageClassName: efs-sc
resources:
requests:
storage: 5Gi
volumeName: efs-shared
Derek Yu
04/19/2023, 1:47 PMHarry Souris
04/19/2023, 6:18 PMHarry Souris
04/19/2023, 6:45 PMtim leonard
04/19/2023, 11:42 PMvalidate download
fails. This usually can be resolved by retrying the download. Is there any way to have a workflow retrigger itself in flyte? Basically if I see an error on validate download
can i just restart the workflow?Nan Qin
04/20/2023, 12:38 AMnode_name
when using with_overrides(node_name=)
in the workflow definition. An error occurred when registering the workflow:
@flytekit.task
def task1(x: int) -> str:
return str(x)
@flytekit.task
def task2(x: str) -> str:
return x
@flytekit.workflow
def baby_training_wf() -> str:
t1 = task1(x=3).with_overrides(node_name='t_1')
return task2(x=t1).with_overrides(node_name='t_2')
error msg:
Failed with Exception: Reason: SYSTEM:Unknown
RPC Failed, with Status: StatusCode.INTERNAL
details: failed to compile workflow for [resource_type:WORKFLOW project:"flytesnacks" domain:"development" name:"tmp.workflow.baby_training_wf" version:"yXlTbWXJo4nsbhCWJe5PGw==" ] with err failed to compile workflow with err Collected Errors: 4
Error 0: Code: NodeReferenceNotFound, Node Id: t_1, Description: Referenced node [t_1] not found.
Error 1: Code: NodeReferenceNotFound, Node Id: t_2, Description: Referenced node [t_2] not found.
Error 2: Code: ParameterNotBound, Node Id: end-node, Description: Parameter not bound [o0].
Error 3: Code: ParameterNotBound, Node Id: t-2, Description: Parameter not bound [x].
Ketan (kumare3)
Blair Anson
04/20/2023, 6:48 AMflytectl create project
command?Boruch Chalk
04/20/2023, 7:21 AMValue error! Received: 403. Request to send data
when submitting a workflow. I have it submitted to a local s3 compatible storage - vast. In the config map, I have set both under 003-storage.yaml
and 010-inline-config.yaml
the required env vars and access key. Any ideas?
003-storage.yaml: |
propeller:
rawoutput-prefix: s3://<bucket-name>/flyte/data
storage:
type: stow
stow:
kind: s3
config:
region: us-east-1
disable_ssl: true
v2_signing: true
endpoint: <vast-endpoint>
auth_type: accesskey
access_key_id: <key_id>
secret_key: <secret>
container: <bucket-name>
010-inline-config.yaml: |
plugins:
k8s:
default-env-vars:
- AWS_METADATA_SERVICE_TIMEOUT: 5
- AWS_METADATA_SERVICE_NUM_ATTEMPTS: 20
- FLYTE_AWS_ENDPOINT: <vast-endpoint>
- FLYTE_AWS_ACCESS_KEY_ID: <key_id>
- FLYTE_AWS_SECRET_ACCESS_KEY: <secret>
Brian Tang
04/20/2023, 8:15 AMnodeName
the pod is currently running on. based on the docs here and the code here, i try to access {{.hostname}}
in the logging configuration, but it’s not populating the template correctly: it’s missing the node name stillStephen
04/20/2023, 11:52 AMEna Škopelja
04/20/2023, 12:31 PM<gs://mybucket/metadata/propeller/myproject-myns-myexecid/n1/data/>
path:
data
| 0/
| | start-node/
| | | 0/
| | | | outputs.pb
|
| inputs.pb
i.e. there's no n0/
directory under data/0/
. When I look into data/0/start-node/0/outputs.pb
it looks to be the correct input to the entire workflow.
I'm running flytekit 1.5.0
and deployed from helm flyte-core:1.5.0
. I tried reverting to 1.4.0
in the core helm chart but it didn't help.
Any idea why this is happening?Amadeusz Lisiecki
04/20/2023, 1:48 PMAlexey Kharlamov
04/20/2023, 2:31 PM@task
def fail_task() -> List[Any]:
return []
@workflow
def fail_workflow():
fail_task()
Using flytekit 1.5.0 and 1.6.0b0 there is error:
Traceback (most recent call last):
File "/opt/venv/lib/python3.8/site-packages/flytekit/exceptions/scopes.py", line 165, in system_entry_point
return wrapped(*args, **kwargs)
File "/opt/venv/lib/python3.8/site-packages/flytekit/core/base_task.py", line 587, in dispatch_execute
raise TypeError(msg) from e
Message:
Failed to convert outputs of task 'pipelines.parallel.fail_task' at position 0:
range() arg 3 must not be zero
SYSTEM ERROR! Contact platform administrators.
In flytekit 1.4.2 the same code works.
Does anyone know how to fix that in newer revisions of flytekit?Bernhard Stadlbauer
04/20/2023, 2:51 PMwait_for_input
.Ahmed Laadraoui
04/20/2023, 4:12 PMadmin:
# For GRPC endpoints you might want to use dns:///flyte.myexample.com
endpoint: dns:///127.0.0.1:8080
authType: Pkce
insecure: true
clientId: <My Client ID here>
logger:
show-source: true
level: 10
then forward the port to 8080 like so:
kubectl -n flyte port-forward service/flyteadmin 8080:81
then I create the project using flytectl create:
flytectl create project --file project.yaml
This is where I am getting the attached logs
After one second I am redirected to this link:
https://localhost:30081/oauth2/authorize?client_id=flytectl&redirect_uri=http%3A%2F%[…]NTRqenJxbWY&state=Z2Zja3JqemRyZ2Q4cDRjd3h3enZkbTh0NnRrbDh2MnQ
Then after couple seconds I see this in the logs and the execution is ended:
Error: authentication error! Original Error: rpc error: code = Unauthenticated desc = token parse error [JWT_VERIFICATION_FAILED] Could not retrieve id token from metadata, caused by: rpc error: code = Unauthenticated desc = Request unauthenticated with IDToken, Auth Error: context was canceled during auth flow
{"json":{"src":"main.go:13"},"level":"error","msg":"authentication error! Original Error: rpc error: code = Unauthenticated desc = token parse error [JWT_VERIFICATION_FAILED] Could not retrieve id token from metadata, caused by: rpc error: code = Unauthenticated desc = Request unauthenticated with IDToken, Auth Error: context was canceled during auth flow","ts":"2023-04-20T16:10:46Z"}
I don’t know if I missed something in the authentication setup.
Your guidance will be very appreciable. Looking forward to it 🙂