-
Notifications
You must be signed in to change notification settings - Fork 94
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
Fix.simulation mode started bug #6351
Fix.simulation mode started bug #6351
Conversation
9453fb9
to
28f4475
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM.
This is a bug (note just a shortcoming in simulation mode) because (as Tim noted) the DB lists the "started" output, but the downstream task is never triggered which is inconsistent.
I did wonder whether this should be fixed by getting the submit code to send the started output like so:
diff --git a/cylc/flow/task_job_mgr.py b/cylc/flow/task_job_mgr.py
index 185966ff1..7a6451605 100644
--- a/cylc/flow/task_job_mgr.py
+++ b/cylc/flow/task_job_mgr.py
@@ -1028,6 +1028,9 @@ class TaskJobManager:
self.task_events_mgr.process_message(
itask, INFO, TASK_OUTPUT_SUBMITTED,
)
+ self.task_events_mgr.process_message(
+ itask, INFO, TASK_OUTPUT_STARTED,
+ )
self.workflow_db_mgr.put_insert_task_jobs(
itask, {
'time_submit': now_str,
But quickly realised why we don't do that:
ERROR - 'execution polling intervals'
Traceback (most recent call last):
...
KeyError: 'execution polling intervals'
This makes Cylc think that there is actually a job running, it tries to setup health checks which fail for obvious reasons.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. Not merging in case @oliver-sanders comment requires a small change?
a8c153e
to
650f987
Compare
@oliver-sanders - final check that you are happy with my response to your comment? |
Not sure which one, but yes I'm happy! |
Fixes a bug discovered by @oliver-sanders in the process of reviewing skip mode work (#6039 (comment)). The bug however, is long-standing (it goes back at least as far as Cylc 8.0.0.
Summary
In simulation mode, job starting is done by the job submission pathway as a side-effect (because in sim mode the two might as well be identical). However, the pathway is not currently spawning children of started.
Example
Run this
Check List
CONTRIBUTING.md
and added my name as a Code Contributor.setup.cfg
(andconda-environment.yml
if present).?.?.x
branch.