[PIPE2D-859] Weekly build fails for w.2021.25 Created: 22/Jun/21  Updated: 29/Jun/21  Resolved: 29/Jun/21

Status: Done
Project: DRP 2-D Pipeline
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Task Priority: Major
Reporter: hassan Assignee: price
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Story Points: 2
Sprint: 2DDRP-2021 A 6
Reviewers: hassan

 Description   

With the current version of master branches, the weekly build fails. See error snippet below:

ERROR: testObjectProducts (__main__.ProductionTestCase_bmn)
Test that object products exist
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/scratch/pprice/jenkins/weekly/2021-06-20/build/stack/miniconda3-4.5.12-1172c30/Linux64/pfs_pipe2d/w.2021.25/python/pfs/pipe2d/weekly/test_weekly.py", line 49, in testObjectProducts
    for target in self.design:
TypeError: iter() returned non-iterator of type 'NotImplementedError'


 Comments   
Comment by hassan [ 28/Jun/21 ]

Problem seems to be a consequence of the changes made in PIPE2D-762. With the changes made there, the return value of pfsDesign.selectFiber(fiberId) can now be a scalar as well as a sequence. A scalar return value its currently not supported in the 2D simulator (specifically lightSource.py), resulting in the failure.

A fix has been pushed to the ticket branch of this issue. But this will be discussed with Paul Price this week to see whether this is better fixed in the datamodel code.

Comment by hassan [ 28/Jun/21 ]

Example simulator run demonstrating the problem:

makeSim --visit 47 --detector r1 --pfsDesignId 1 --exptime 1800 --type object

Run against weekly data (see eg /scratch/hassans/weekly/PIPE2D-859).

Comment by price [ 29/Jun/21 ]

The weekly doesn't use the simulator (only its outputs), so we need to fix this in datamodel. I removed some functionality recently that apparently is used in testing the weekly, so it needs to go back in.

Comment by price [ 29/Jun/21 ]

But go ahead and merge the changes to drp_instmodel too: we're going to need them.

Comment by hassan [ 29/Jun/21 ]

Yes you are right, the issue with the simulator has nothing to do with the weekly builds. The simulator code has broken following the changes to the datamodel, but this is different to the problem with the weekly.

I will merge the simulator problem as a separate ticket.

Comment by hassan [ 29/Jun/21 ]

Changes made by price in datamodel to fix the failing weekly look fine to me.

Comment by price [ 29/Jun/21 ]

Merged.

Generated at Sat Feb 10 15:59:00 JST 2024 using Jira 8.3.4#803005-sha1:1f96e09b3c60279a408a2ae47be3c745f571388b.