| 25 Mar 2024 |
 mpysik | * I just noticed that the way I rewrote this function doesnt even use the tag_name argument, which seems wrong | 13:40:40 |
| mpysik | (the one I posted is the original, without my rework) | 13:41:02 |
 lmjachky | DeclarativeArtifact can be used later in the staging when running the sync pipeline; it is used for initializing artifacts that can be downloaded on-demand | 14:00:19 |
| lmjachky | and it can be also used when creating publications | 14:00:30 |
| lmjachky | since we have our customized Handler (registry.py), I THINK we do not need to worry much about URLs for declarative artifacts | 14:01:33 |
| lmjachky | and the URLs is the point where the tag name was used previously | 14:01:45 |
| lmjachky | mpysik ^ | 14:01:49 |
| lmjachky | also, we want to have manifests in place, so no on-demand downloading should be incorporated; thus, no declarative artifacts are needed in this case | 14:02:32 |
| lmjachky | * also, we want to have the content of manifests in place, so no on-demand downloading should be incorporated; thus, no declarative artifacts are needed in this case | 14:02:40 |
| lmjachky | because everything is set on the object/model level | 14:02:53 |
| lmjachky | does it make sense? | 14:02:58 |
| 28 Mar 2024 |
| mpysik | Hey, I'm not really able to progress further without a way to test what I write - the tests for registry.py etc. call endpoints, so if the test fails in pytest, it only tells me that it returned http status 500 or other things in similar fashion, but I don't get the python trace to see what's wrong | 09:22:49 |
| mpysik | compose logs are also not really helping in this | 09:23:04 |
| lmjachky | I run oci-env as "oci-env compose up" and I can see the logs in real time; tracebacks for API calls should be visible there; for tasks, you may want to preview tasks in the REST API | 09:32:21 |
| lmjachky | can you elaborate on what errors do you see and where, mpysik? | 09:32:40 |
| mpysik | In reply to @lubosmj:matrix.org
does it make sense? so I guess the tag was passed just to associate it with the artifact...which we don't need anymore since there is no artifact | 09:33:41 |
| lmjachky | How do you preview logs/tracebacks? | 09:36:29 |
| mpysik | 
Download image.png | 09:37:15 |
| mpysik | I get stuff like this very often, logs show no traceback | 09:37:25 |
| mpysik | that was locally, so it doesnt matter that much | 09:39:44 |
| mpysik | 
Download image.png | 09:39:48 |
| mpysik | but this is what the github pipeline shows, and no traceback in logs either | 09:40:00 |
| mpysik | so in this case, we got 404 not found, but no traceback on what's happened inside the code | 09:41:22 |
| lmjachky | aha, I see | 09:42:14 |
| lmjachky | the idea of this test is to check if the redis cache is correctly invalidated/refreshed when accessing different content units after changing some of the entities | 09:43:14 |
| lmjachky | what is the expected value here and what is being returned by pulp? | 09:44:02 |
| lmjachky | we may want to hop on a call because we discovered some issues with these tests in the past with hyagi | 09:44:22 |
| lmjachky | * we may want to hop on a call because we discovered some issues with these tests in the past with hyagi as we removed the schema conversion | 09:44:53 |
| mpysik | In reply to @lubosmj:matrix.org
we may want to hop on a call because we discovered some issues with these tests in the past with hyagi as we removed the schema conversion sure | 10:01:25 |
| 4 Apr 2024 |
|  dkliban joined the room. | 13:18:37 |
