Open casparvl opened 2 days ago
Full traceback:
Traceback (most recent call last):
File "/usr/lib64/python3.6/runpy.py", line 193, in _run_module_as_main
"__main__", mod_spec)
File "/usr/lib64/python3.6/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/home/bot/eessi-bot-software-layer/eessi_bot_job_manager.py", line 733, in <module>
main()
File "/home/bot/eessi-bot-software-layer/eessi_bot_job_manager.py", line 724, in main
job_manager.process_finished_job(known_jobs[fj])
File "/home/bot/eessi-bot-software-layer/eessi_bot_job_manager.py", line 570, in process_finished_job
update_comment(int(pr_comment_id), pull_request, comment_update)
File "/home/bot/eessi-bot-software-layer/tools/pr_comments.py", line 136, in update_comment
tries=5, delay=1, backoff=2, max_delay=30)
File "/home/bot/.local/lib/python3.6/site-packages/retry/api.py", line 101, in retry_call
return __retry_internal(partial(f, *args, **kwargs), exceptions, tries, delay, max_delay, backoff, jitter, logger)
File "/home/bot/.local/lib/python3.6/site-packages/retry/api.py", line 33, in __retry_internal
return f()
File "/home/bot/.local/lib/python3.6/site-packages/github/IssueComment.py", line 131, in edit
"PATCH", self.url, input=post_parameters
File "/home/bot/.local/lib/python3.6/site-packages/github/Requester.py", line 355, in requestJsonAndCheck
verb, url, parameters, headers, input, self.__customConnection(url)
File "/home/bot/.local/lib/python3.6/site-packages/github/Requester.py", line 378, in __check
raise self.__createException(status, responseHeaders, output)
github.GithubException.GithubException: 422 {"message": "Validation Failed", "errors": [{"resource": "IssueComment", "code": "custom", "field": "body", "message": "body is too long (maximum is 65536 characters)"}], "documentation_url": "https://docs.github.com/rest/issues/comments#update-an-issue-comment", "status": "422"}
https://github.com/EESSI/software-layer/pull/726 fixes the issue that the overview doesn't correctly find the accelerator builds and just interprets all files as "other stuff". However, it may still be good to also implement a check on the length of the entire message that is used for the Github comment.
This build of CUDA was successful, but never reported back. In the bot logs, Kenneth saw
This is probably due to the fact that a) CUDA installs a lot of files and b) the prefixes are still quite long (e.g. for the UCX-CUDA module, the prefix was
accel/nvidia/cc80/modules/all/UCX-CUDA/1.14.1-GCCcore-12.3.0-CUDA-12.1.1.lua
), since we should still strip theaccel/nvidia/cc80/
part.It's probably good if the bot limits the message body in case it is too long. It's better to hard cut it, and add a final line that says the message was cut short because it was too long, then to get no response at all. Alternatively (but this would have to happen in the
software-layer
, we should limit the size of theartifacts
list, so that things that come after it (status, test result, etc) would still be displayed...