Github user jpeach commented on the pull request:
https://github.com/apache/trafficserver/pull/44#issuecomment-39381147
I'm not confident that this change is correct. The root cause of the
assertion seems to be that `cache_info.transform_store` is not handled the same
way as `cache_info.object_store`. In `HttpTransact::handle_transform_ready`, we
know that the `transform_response` is valid, so that might be a reasonable
place to set the response into `cache_info.transform_store`.
I would also like to understand why
`HttpTransact::set_headers_for_cache_write` cares about caching a negative
response at all. I *think* that this is all about setting up a special response
to write to cache when caching a negative response and being careful not to
trash that with a different response header.
Finally, it sounds like it should be simple to write a test for this in the
`ci/tsqa` harness using the null transform example plugin and curl.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
