These are chat archives for whereat/contrib

Apr 2016
Bradley Menchl
Apr 11 2016 02:29
I'll look into Espresso, but I also found a mention in the react native subreddit
apparently fb uses Selendroid (which I believe Appium uses..) and it has the ability to find by tag name, which is what Android TestID sets.
Apr 11 2016 02:48
Kamal Marhubi
Apr 11 2016 04:09
@aguestuser how is the CI stuff going? I got a bit busy and didn't get around to doing the concourse tutorial
Apr 11 2016 05:36
so glad you asked! its ALMOST great!
once it's installed, it's AMAZING -- declartive manifests for pipelines that you can edit and deploy from the (really great) CLI, very user-friendly web UI -- definitely something i want to use!
only problem is the install is a bit tricky
in the getting started tutorial you work off a local vagrant box, which is easy
but in prod, the vagrant box setup (which they call bosh-lite) makes github OAuth fail -- as documented in whereat/whereat-ci@7628960
so i spent the weekend figuring out how to deploy it to a full BOSH cluster, and documented the process in the README for our ci repo:
Apr 11 2016 05:42
trying to run with this configuration produces an odd error message:
resource script '/opt/resource/check []' failed: exit status 1

failed to ping registry: 2 error(s) occurred:

* ping https: Get net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)
* ping http: Get net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)
as best i can tell, this is because the concourse worker trying to run the pipeline can't connect to docker hub
someone had a similar issue here: concourse/concourse#347
and some general discussion about why sometimes boxes can't connect to docker hub, here:
(the "hello world" pipeline consists of pulling down the base ubuntu image from docker hub, then echoing "hello world" in bash)
the pipeline is failing when it tries to pull that image...
feeling a bit drained from a weekend of toolchain wrangling so going to sleep on it for the night and try to make sense of things with @alxmrtn tomorrow.
Apr 11 2016 05:48
a bit new to AWS, so have a feeling that it's something about how the Virtual Private Cloud subnet is configured causing the problem, but i've looked very closely at all the documentation specifying how you're supposed to set bosh up for concourse on AWS (and bosh up on AWS more generally), and can't find anything i haven't done. so... a bit stumped for the moment... but not defeated yet!
Paul Nasrat
Apr 11 2016 13:45
Sorry I've been focussed on RC - I'm hoping to spend some time Friday PM catching up on whereat
Apr 11 2016 16:48
@pnasrat cool! :D
ps: apropos of the conversation about the signal protocol earlier -- saw this on today: