So I'm more and more puzzled on how to deploy a Prisma instance as the docs are out of date or erroneous (both the DO, either manual or docker-machine are failing at one point or another, on Prisma 1.6)

Which version are you trying to deploy?

which version of what?

Prisma ... 1.6?

yes

I'm running 1.6 on DO and I used the docs. What problems are you having?

“Cannot create container for service prisma-db: Conflict. The container name /prisma-db is already in use by container XX. You have to remove (or rename) that container to be able to reuse that name”

I know next to nothing about docker (and try to keep it that way). I've had that error locally, though, and was always able to restart docker to get it to go away. I assume you tried that and the error persists?

the Docker app itself?

Yes

I'll try that

Well "restart docker" isn't exactly great help. 🙂 I suspect trouble shooting it as a Docker issue is the way to go

try to kill & remove containers (might incur data loss)

@veksen I second Nilan’s statement, more precisely try

docker kill $(docker ps -aq) && docker rm $(docker ps -aq)

. Caution this permanently removes all containers from your system, so if you have more containers running you can selectively delete containers with

kill / rm

.

that works! thank you so much!

Glad it works.

well - back to not working. same error, also did a clean start by killing and removing all containers (within docker-machine + docker). very puzzled

Did it just stop working, or did you do anything prior that lead to the container failing, like deploying new containers?

I mean, I started the tutorial again to make sure I got everything right, and it started failing when trying to enable cluster authentication, following the steps again (https://www.prisma.io/docs/tutorials/cluster-deployment/digital-ocean-(docker-machine)-texoo9aemu)

so you can run the same steps to get unblocked again 🙂

I did run the same steps several times, I spent a full day on the tutorial.

with same steps I mean

docker kill xyz

and

docker rm xzy

🙂 I am not sure at which specific step the tutorial it causes this issue

Alright what I'll do is... Kill any existing config/docker-machine + docker instances and start from a blank state, and log what goes wrong. I'll report back. Thanks for the follow-up

any news @veksen? 🙂

on it today 🙂 needed a long weekend

same! 😄

alright so I have a full step-by-step repro of my attempt from a blank state: https://gist.github.com/veksen/c45b4e554bcd773ca64f44ea098b94c0

ok, interesting, thanks for providing that! note that anything surrounding

~/.prisma/config.yml

are unlikely to be changed again, as it is supposed to be not needed anymore: https://github.com/prismagraphql/prisma/issues/2356

the interesting/weird part is the cluster inside

~/.prisma/config.yml

is changed to

local

, but I feel like it's just a symptom from the error

that might be a proper bug, but as I mentioned due to the changes around

config.yml

it is not applicable anymore for recent versions

ya I'm aware.. I would rather switch to 1.8. I'd gladly ginnea pig a DO deployment for 1.8 - My docker experience is limited however