-
Notifications
You must be signed in to change notification settings - Fork 523
Make rabbitmq host in queue-master configurable just like in shipping #17
Conversation
The coveralls.py check failed, but I believe that is because the COVERALLS_TOKEN string was empty for me and that this might always occur for forks since my repository does not have the actual token that you guys have in the main microservices-demo repository. I did just log into travis-ci.org with my GitHub password. Perhaps that will enable the COVERALLS_TOKEN to be set correctly in the future? Is there a way to re-run the test? |
I'm going to close and resubmit this in the hope that the Travis tests will run correctly since I have now linked my Travis CI account to my GitHub account. |
I also logged into coveralls.io and enabled this repository there. I will close and re-open and see if that successfully builds. |
Re-opening to rerun test |
No matter what I do, it seems that the COVERALLS_TOKEN string is blank, causing the coveralls.py test to fail. This might be typical with forked repositories according to https://forum.micropython.org/viewtopic.php?t=3007. But maybe I just haven't set things up the right way so that COVERALLS_TOKEN is populated correctly when I submit a pull request here. Please advise. |
Any chance of getting this committed early next week? This is the last thing I need to enable running the entire Socks Shop app on Apcera with docker images instead of having to download and modify source code from GitHub. We would like to show this at DockerCon in a few weeks. Thanks, |
Re-opening after turning switch on for my repository on travis-ci.org to see if the coveralls test will pass . I suspect not since I probably still won't have access to your coveralls token. |
This time, closing and re-opening did not even cause Travis to run the test again. |
I pushed another commit to comment out the spawning of Docker, in line with #3. Hoping that closing and re-opening will pull in that commit. |
Commenting out docker.init() and docker.spawn() lines is ok, but commenting out DockerSpawner was not since that leaves the ShippingTaskHandler without any bean that qualifies as autowire candidate for the Shipment dependency. Now the test fails in coveralls test again. Note that the prior test cannot connect to rabbitmq server, but none is spun up during the test. So, that is presumably expected. If the reviewers object to my commenting out the docker.init() and spawn() lines, I can do another commit and restore them. |
Hi @rberlind, thanks for your PR! I'm sorry but we won't have time to review it before DockerCon next week, however please let me know if I can help you building custom images so that you can demo from your fork. |
Thanks Luke, |
I went ahead and removed my commenting out of the docker spawning in the hopes that this might make it easier to merge my pull request with less review since the only change is now to make the rabbitmq host configurable exactly in the same way as is done in the shipping service. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It appears to pass but you have no access to Coveralls token which is throwing an error. Otherwise it appears to be fine.
I'm submitting this pull request in connection with issue microservices-demo/microservices-demo#686 and #16 which contains excerpts from the first.
The code changes are copied directly from shipping, so should not cause any problems.
I also just successfully ran ./test/test.sh unit.py and ./test/test.sh component.py against the modified code in my cloned repository and both were successful, although the latter did give exception "org.springframework.amqp.AmqpIOException: java.net.UnknownHostException: rabbitmq" because there was no rabbitmq host on which rabbitmq could be listening. But I think that's ok.