What are the disk space limitations for the build?

[Ben]

This question is in reference to Atlassian Documentation: Limitations of Bitbucket Pipelines (beta)

In pipelines, are there limits on how much disk space you can use during the build, and what is that limit?

[Sten Pittet]

Hey Ben, just to get more context would you mind telling us how much space your builds need on average?

Thanks

[Sten Pittet]

Hi Ben,

We recommend to keep the disk space required to run the pipeline under 5GB. We're monitoring usage during the beta to inform better what limits we will have when we make Bitbucket Pipelines available to all customers.

Thanks,

Sten

[Ben]

We have one project (sort of a custom packet-based db) that has tests that generate some pretty large files (that get deleted after test), but there's all much less than 5GB - perhaps 1GB of files at most.  Just investigating some test failures there, and wanted to make sure we not hitting some limits.

For reference, we added some diagnostics to our build steps - so I don't think disk space is the issue.  We have noticed weird behavior of the docker filesystem though on other tests (we're pretty new to docker - issues with filesystem not syncing quickly at all, even with explicit sync's in our code), so perhaps we're seeing something like that in these tests too.

```

df -h
+ df -h
Filesystem Size Used Avail Use% Mounted on
none 197G 65G 125G 35% /
tmpfs 16G 0 16G 0% /dev
tmpfs 16G 0 16G 0% /sys/fs/cgroup
/dev/xvda1 197G 65G 125G 35% /root/.m2
shm 64M 0 64M 0% /dev/shm

```

I'm trying to recreate failures on local docker, but get different results.