When SSHed into a container VCAP_APPLICATION is missing instance_index when == 0

[youngm]

If I ssh into the 0 instance of a container, the VCAP_APPLICATION environment variable is missing the instance_index property. Though the environment variable in the running application does have instance_index=0.

This is obviously a low priority issue but thought I'd surface it anyway and this project seemed like a decent place to start.

[cf-gitbot]

We have created an issue in Pivotal Tracker to manage this:

https://www.pivotaltracker.com/story/show/150829908

The labels on this github issue will be updated when the story is started.

[emalm]

Hi, @youngm,

Thanks for pointing out the discrepancy! That's another consequence of the SSH session not having exactly the same environmental setup as the app start command by running through the launcher executable appropriate for that app lifecycle type (buildpack, Docker, or Windows 2012). That launcher injects the instance ID and index into the VCAP_APPLICATION JSON payload from the values of the index- and ID-specific environment variables. It looks like we do document that as an environmental difference in https://docs.cloudfoundry.org/devguide/deploy-apps/ssh-apps.html#ssh-env, and if we prioritize work to fix those differences I would expect this one to get corrected then.

Best,
Eric, CF Diego PM

[youngm]

@emalm Nice job documenting it. I guess now I'll have to check documentation too when I think I've found a bug. :) Thanks.

[emalm]

Thanks again, @youngm !