When uptimer fails to push its app - it should not run its `while`

[zachgersh]

Had this happen today, our CF is misconfigured and our app failed to push. Even though the app broke uptimer still went into its run of while. I guess I could cancel but it would be nice for this to bail our early.

[cf-gitbot]

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

https://www.pivotaltracker.com/story/show/150906007

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

[dsabeti]

Hey @zachgersh, this seems like a reasonable request, but we actually need a flag to keep the existing behavior. We use uptimer in our Concourse tasks to do a bosh deploy, so we need a way to make sure that uptimer completes the while command even if the initial push fails.

I think it makes sense for your proposed behavior to be the default, and then to add a --finish-command flag or something like that, which would always finish the while command.

[sunjayBhatia]

I would like to bump this as a requested feature for uptimer

If the uptimer testing app fails to push (due to a flake, etc.) then I would rather the command under test does not run so the side effects of the command (a bosh deploy) are not carried out, so as not to cause a deployment for example to continue without uptime testing

[heyjcollins]

What do you think about adding a retry routine + the optional --finish-command flag (with default being cease command if setup fails)?

cc @Syerram

[sunjayBhatia]

@heyjcollins I'm on board with that!