Closed mesmith75 closed 2 months ago
This is looping over the subjobs. When it gets to the last one, then it also updates the master job status. There is another updateStatus
two lines later for the master job which tries to update the already updated master status and gives the error. It is better to do this than remove the second one, as the second one is also used for jobs without subjobs.
You are right though, I'll check again to be sure it is doing exactly what I think it should be
You are right - this wasn't the real issue. The problem is that the master job ends up in failed
once the last subjob is killed but then the following master job update tries to put it into killed
, causing the issue.
I am not sure what is best - have the master job switch to killed
or stay at failed
.
This is slightly less elegant, but more precise for what we want
Fixes #2358