CodeCatalyst Dev Environment Large Repository Start Problem

0

I have for the last few days trying to get a Dev Environment to work that uses a custom image and presumed my problems with it was due to that customization.

However the problem seems to originate that starting a Dev Environment with a large repository (~2G of source and ~3.5G with git files included) results in the start of a new dev environment to fail.

Is there a way to extended how long the start waits till it times out on startup?

profile picture
asked 4 months ago114 views
2 Answers
0

Hey there

did you try to use the DevEnvironment in the browser or using the "remote" option? https://docs.aws.amazon.com/codecatalyst/latest/userguide/devenvironment.html

what's the "size" of the dev environment that you are using? (https://docs.aws.amazon.com/codecatalyst/latest/userguide/devenvironment-limits.html)

Did you try to upgrade to the standard tier and use a " bigger" Dev Environment? Does the problem persist?

Regards Johannes

profile picture
answered 4 months ago
0

Hi Johannes,

Thank you very much for your reply.

As to your first point, yes I could simply check out the repo and work on it locally the way I have always worked with repos in the past, however I am trying to convince the web developers/designers I work with to start using version control and in general a more structured approach to work those projects and the point and click nature of CodeCatalyst's dev environments would be a massive selling point (in particular having a local dev webserver setup the way it is on the production server ... or at least close enough).

I was even considering, but have not tried that, to have an empty "create new dev environment branch" in the repo, followed by a switch to the actual code branch.

As to your second point I originally started with creating a docker image derived from the universal image which ballooned quickly to over 9G. I consequently created a slim line image based on the Amazon Linux 2023 base which in my local docker reports as 288M which shows as 127M in ECR (presumably ECR's size is the compressed storage size vs the the net expanded size in my docker).

The repo I am working with has a approximately 2GB source code size and including the git files (the content of the .git directory) is about 3.5G in size.

Judging by the following error message when trying a VSCode Dev Environment (which gives me access to the log compared to Cloud9), I would judge that there is some sort retry after a timeout which is exceeded when the initial git clone takes too long:

message: 'Cannot start Dev Environment because start process is still going on',

I have not yet tried to go larger with my instances, but I will give this a try, even so, I am not sure if adding more CPU/Memory will really speed up git sufficiently.

With best regards, Roland

profile picture
answered 4 months ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions