Details
-
Type:
Story
-
Status: Won't Fix
-
Resolution: Done
-
Fix Version/s: None
-
Component/s: faro
-
Labels:None
-
Team:DM Science
-
Urgent?:No
Description
Copied from Lee's comment on related ticket:
Testing this on the tiger machine here at Princeton, I get the following values:
result.measurement = 9.55842919753633e-05
expected = 9.517637187446586e-05
With such a large gap, I'm not sure that anything like assertAlmostEqual will ever be appropriate here, and perhaps we'll need to consider something else to allow this test to pass?
Attachments
Issue Links
- relates to
-
DM-31611 Improve reproducibility in faro ellipKPM test_te1
- Done
Running a freshly installed Springdale Linux in a VM and testing with a freshly installed binary release of lsst_distrib (w_2021_36) does not reproduce the problem. Would be interesting to investigate what's the different run-time setup on tiger causing this.