Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Optimize sharness tests #4961

Open
2 of 5 tasks
magik6k opened this issue Apr 22, 2018 · 2 comments
Open
2 of 5 tasks

Optimize sharness tests #4961

magik6k opened this issue Apr 22, 2018 · 2 comments
Labels
help wanted Seeking public contribution on this issue kind/test Testing work

Comments

@magik6k
Copy link
Member

magik6k commented Apr 22, 2018

Some sharness tests take way more time to complete than they have to.

Results from https://ci.ipfs.team/job/IPFS/job/go-ipfs/job/PR-4959/7/testReport/sharnessLinux/ (sort by duration) / #4959

  • - t0181-private-network / 12 - try connecting node in public network with priv networks - should timeout way sooner
  • - t0130-multinode - adds a tree of 5k files multiple times, might be a bit overkill, at least of something that isn't a benchmark
  • - t0250-files-api is huuuuge, would be good to split into some separate tests to make use of make parallelism
  • - t0260-sharding operates on quite large directories, not sure if we can do much about it
  • - t0040-add-and-cat does some stuff on large files, but majority of the time spent there seems to be spent in the ton of tests there, might be worth to split too
@magik6k magik6k added help wanted Seeking public contribution on this issue kind/test Testing work labels Apr 22, 2018
@kevina
Copy link
Contributor

kevina commented Apr 25, 2018

I am probably responsible for the huge number of tests in t0250-files-api and t0040-add-and-cat. I am skeptical on the benefit of splitting these tests as they are just one of many shareness tests that need to be run, so the probability that they will increase parallelism is low. Nevertheless, if we do split them up I would like to take lead on that.

@Kubuxu
Copy link
Member

Kubuxu commented May 7, 2018

@kevina it will decrease the duration of the tests significantly as these are the tests we end up waiting for at the end of the run.

If we split them in half (online and offline) it is quite possible that they will run in parallel with other tests instead of being the only two tests running at the end.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
help wanted Seeking public contribution on this issue kind/test Testing work
Projects
None yet
Development

No branches or pull requests

3 participants