More accurate metrics for HTTP redirect responses from Jersey

[tinolazreg]

Fixes #3327

[pivotal-cla]

@tinolazreg Please sign the Contributor License Agreement!

Click here to manually synchronize the status of this Pull Request.

See the FAQ for frequently asked questions.

[shakuzen]

Thanks! It looks good to me. Could we add a test for the case when there is no matched route?
Also, we will need the CLA signed to accept the contribution.

[pivotal-cla]

@tinolazreg Thank you for signing the Contributor License Agreement!

[tinolazreg]

Thanks! It looks good to me. Could we add a test for the case when there is no matched route? Also, we will need the CLA signed to accept the contribution.

Great! I have signed the CLA now. I tried creating a case for the test that you mentioned, but found it difficult to create a similar test with a HTTP redirection without a matching route, that ends up in the JerseyTags.

[shakuzen]

I tried creating a case for the test that you mentioned, but found it difficult to create a similar test with a HTTP redirection without a matching route, that ends up in the JerseyTags.

Thank you for trying. I tried a bit myself and also struggled to come up with something. I don't like putting something in that isn't tested, but the logic makes sense and is similar to instrumentation we have elsewhere.

I've rebased the pull request on the 1.8.x branch so we can get this fixed in upcoming maintenance releases. The more I've thought about it, the more the previous behavior was more bug-like. Once merged, snapshots will be available if you would like to test things out with your application before the release.

Thank you for reporting and fixing this.

[tinolazreg]

I tried creating a case for the test that you mentioned, but found it difficult to create a similar test with a HTTP redirection without a matching route, that ends up in the JerseyTags.

Thank you for trying. I tried a bit myself and also struggled to come up with something. I don't like putting something in that isn't tested, but the logic makes sense and is similar to instrumentation we have elsewhere.

I've rebased the pull request on the 1.8.x branch so we can get this fixed in upcoming maintenance releases. The more I've thought about it, the more the previous behavior was more bug-like. Once merged, snapshots will be available if you would like to test things out with your application before the release.

Thank you for reporting and fixing this.

Thank you! Glad to be able to contribute.