-
Notifications
You must be signed in to change notification settings - Fork 48
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ResourceManager.cluster_application_kill() is not working #75
Comments
PR #77 should resolve the application submission issue if anyone wants to try the included python code to reproduce the kill issue. |
Brief look over stacktrace seem to refer to JAXB mapping of input JSON body. Perhaps headers for application/json are not passed for some reason? So body is not considered proper json. |
Can you also confirm your version of |
@dimon222 - thanks for the response. Yeah, one of the google hits indicated a missing application/json entry in the header, but I confirmed that we're sending that into the requests method. Also, given the curl and REST clients also hit the same endpoint, indicates there must be something getting whacked in This issue reproduces using requests versions of |
@kevin-bates try in these arguments to change |
I've tested it. It works using ^ what i mentioned there. I haven't checked the other methods, but its possible that other |
Excellent! Thank you for the PR. |
In troubleshooting issues in another repo, I found that
ResourceManager.clsuter_application_kill()
is resulting in500
status code with the following message: "Response finished with status: 500. Details: "However, using CURL or REST clients that issue calls, result in the expected results...
curl -X PUT -H 'Content-Type: application/json' -i 'http://yarn-eg-node-1.fyre.ibm.com:8088/ws/v1/cluster/apps/application_1574459664527_0116/state' --data '{"state":"KILLED"}'
All portions of the request appear to match what is used in curl, so I'm hoping someone has ideas about this.
I've also included a python file that submits an application, checks its state and attempts to kill via the resource manager. Upon determining the resource manager fails, it issues the kill call using curl, then confirms the state is KILLED.
((Please note that this python test uncovered an issue with
cluster_submit_application()
in which the response of no content could not be processed viaresponse.json()
. I will be submitting a PR for that issue shortly.))Looking at the YARN logs, they contain the following traceback ...
The text was updated successfully, but these errors were encountered: