Fixing remote test failures #2589
Conversation
Codecov Report
@@ Coverage Diff @@
## main #2589 +/- ##
=======================================
Coverage 64.19% 64.19%
=======================================
Files 130 130
Lines 16881 16881
=======================================
Hits 10836 10836
Misses 6045 6045 📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
|
Some of these are duplicates for the previous pr, could you maybe rebase on top of that? |
|
@bsipocz Sigh, didn't notice your PR. Should I rebase, or just close this one and work on yours? |
|
I would go for a quick merge on that one and a rebase here. We never get rid of all of the failures in one PR, so incremental is better for this than keeping things open for a long time. |
|
@bsipocz Alright, poke me when you've merged yours and I will rebase. |
|
@ceb8 - it's merged. |
| with pytest.raises(AttributeError): | ||
| alma.is_proprietary('uid://NON/EXI/STING') | ||
|
|
||
| @pytest.mark.skipif("SKIP_SLOW") |
There was a problem hiding this comment.
It would be better to use the slow marker introduced in pytest-astropy version 0.10.0. See also #2420.
| esa_hubble = ESAHubble() | ||
| c = coordinates.SkyCoord("00h42m44.51s +41d16m08.45s", frame='icrs') | ||
| temp_file = self.temp_folder.name + "/cone_search_m31_5.vot" | ||
| temp_file = self.temp_folder.name + "/cone_search_m31_5.vot.gz" |
|
@eerovaher This PR is marked as draft because it is not ready for a detailed review, please hold off on detailed comments. |
ceb8
force-pushed
the
remote_failures
branch
from
1eb9c7a to
13a157a
Compare
|
@ceb8 - in the OP you mention unclosed sockets. I believe I added those to the filtered exceptions already, do you still see them? (and they come from all over the place, bs4, standard library parts, pyvo, astropy, so I don't think we can do much about them this hight up the foodchain) |
I don't see this one being anywhere this slow (but alma tests are in fact somewhat flaky and we have a few issues open that they are the cause for the GHA timeouts we sometimes see), neither in cron nor locally (https://github.com/astropy/astroquery/actions/runs/3495228393/jobs/5851757125#step:5:23845). So I would suggest opening a separate issue for it, and tag Adam and Adrian as something may be genuinely wrong. |
ceb8
force-pushed
the
remote_failures
branch
from
d0837d3 to
0301782
Compare
|
@ceb8 - This has evolved a lot due to changes in |
ceb8
force-pushed
the
remote_failures
branch
from
59f36d8 to
908f34b
Compare
|
@bsipocz done |
|
Thanks @ceb8! |
3 participants
I'm starting this draft now so there is a place to discuss the remote test failures and solutions.
Fixes
test_download_producttest_query_target_errorfail because the wrong exception is raised. I could not find an easy solution, so I commented out the tests and captured the issue here: ESA JWST remote test failues #2590test_get_imagestest to be a faster running query, and marked it to ignore the FITS file warning (this has nothing to do with the functioning of the module)