You're on your way to the next level! Join the Kudos program to earn points and save your progress.
Level 1: Seed
25 / 150 points
1 badge earned
Challenges come and go, but your rewards stay with you. Do more to earn more!
What goes around comes around! Share the love by gifting kudos to your peers.
Keep earning points to reach the top of the leaderboard. It resets every quarter so you always have a chance!
Join now to unlock these features and more
I would like to configure docker memory limit differently depending on the step I run.
In fact I have a step that perform tests and use a lot of memory using docker-compose, but some other steps that just build the docker image.
For instance one step that use docker with 1GB and another one with 6GB to hold my several services
Is it possible to override docker memory limit for a specific step ?
We have the exact same requirement as @Victor Fleurant . We are running our tests in docker-compose and get "Container 'docker' exceeded memory limit." some times. It's really annoying and breaks the whole CI process.
Just need the "Docker" service in the Test step to have more memory, but there's no way to define that other than adding a memory limit in "definitions", which affects other steps using the "Docker-in-Docker" service.
Any chance of making a feature request for this?
Agree, this is really annoying and not clear.
As a compromise workaround i found that you can redefine (extend) docker service and use it in your step instead. That way your definition for it does not affect basic docker service and you don't have to adjust the size of other steps.
The drawback is that you cannot simply use `caches: -docker` anymore. May be there is a workaround for this also...
Al always, atlassian's documentation is just not detailed enough.
name: build docker image
name: update image on test server
- pipe: atlassian/ssh-run:0.4.0
From what I remember you can control the memory limit first of all by choosing between two options, the standard one (4GB) and then there is one to double the resources (8GB).
If you want to bring memory limit down (from either of those two), you have to add services. Each service divides the overall limit. E.g. Having three services brings down the limit for the pipeline script itself from 4GB to 1GB.
Going from 8GB to 6GB does not look to be mathematically possible this way, only from 8GB to 4GB (one service) but then you could just run with 4GB as the default.
Running docker rootless inside a pipeline would not add more options here as it is limited on limiting resources:
Currently, rootless mode ignores cgroup-related docker run flags such as --cpus and memory.
However, traditional ulimit and cpulimit can be still used, though they work in process-granularity rather than in container-granularity.
E.g. see ulimit and sysctl (www.LinuxHowtos.org) and the administration and configuration guide of your choice.