Hi All, I Got An Issue With The Pipliene Before Create Launching The Next Steps What Can Be Help To Solve This? File "/Home/Core/.Clearml/Venvs-Builds.1.2/3.8/Lib/Python3.8/Site-Packages/Clearml/Automation/Controller.Py", Line 795, In Start Self._Start(

Answered

Hi all,
I got an issue with the pipliene before create Launching the next steps
what can be help to solve this?

File "/home/core/.clearml/venvs-builds.1.2/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 795, in start
self._start(
File "/home/core/.clearml/venvs-builds.1.2/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 1197, in _start
self._prepare_pipeline(step_task_completed_callback, step_task_created_callback)
File "/home/core/.clearml/venvs-builds.1.2/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 1225, in _prepare_pipeline
self._deserialize(pipeline_dag)
File "/home/core/.clearml/venvs-builds.1.2/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 1453, in _deserialize
self._nodes = {
File "/home/core/.clearml/venvs-builds.1.2/3.8/lib/python3.8/site-packages/clearml/automation/controller.py", line 1454, in <dictcomp>
k: self.Node(name=k, **{kk: vv for kk, vv in v.items() if kk not in ('job_id', )})
TypeError: init() got an unexpected keyword argument 'relaunch_on_instance_failure'

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					GaudyCoyote83
				
					0
					 × 1

Votes Newest

Answers 8

Hi GaudyPig83 , when is this happening?
Do you have a code snippet that reproduces this?

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					CostlyOstrich36
				
					0

I do import for exist code in python that run remote pipline as part of clearml agent

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					GaudyCoyote83
				
					0
					 × 1

GaudyPig83
I think there is some mismatch between the code creating the pipeline and the actual Task?! Could that somehow be the case? "relaunch_on_instance_failure" is a missing argument somehow

can you try to launch the entire Pipeline with the latest RC ?
pip3 install clearml==1.7.3rc0

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					AgitatedDove14
				
					0
					 × 1

however when i change the flag to run locally - it's run without errors

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					GaudyCoyote83
				
					0
					 × 1

'relaunch_on_instance_failure'

This argument is Not part of the Pipeline any longer, are you running the latest clearml python version?

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					AgitatedDove14
				
					0
					 × 1

GaudyPig83 AgitatedDove14 I'm experiencing the same issue. was this resolved?
BTW, I'm running a self-hosted server with latest versions of server, agent and clearml

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					HugeArcticwolf77
				
					0
					 × 1

HugeArcticwolf77 I think this issue was resolved with the latest version 1.8.0, can you try to rerun the entire pipeline with the latest version?

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					AgitatedDove14
				
					0
					 × 1

I did. Runing clearml==1.8.0, clearml-agent==1.4.1 and clearml-server=1.7.0

edit: I reran the script from terminal instead of the GUI and it works. thanks AgitatedDove14 !

  				
Posted 
	2 years ago

					More
				  		
  Report
		
					HugeArcticwolf77
				
					0
					 × 1

Write your answer

1K Views

8 Answers

2 years ago

one year ago