Issue #14917 resolved
Morten Meiling Olsen
created an issue

Well, pretty much as the title says. It would be extremely helpful to be able to retry a single step in a long pipeline...

Official response

  • Aneita Yang staff

    Hi everyone,

    We're super excited to announce that the ability to rerun failed steps is now available to all Pipelines users. This lets you save both time and build minutes. For more information about step rerun, check out our documentation.

    If you are interested in the ability to rerun successful steps for the purpose of rolling back changes, I encourage you to vote for / watch issue #16604 which is an open issue to track that functionality separately. If you have any other suggestions or requests that weren't covered as part of this feature, please raise a new feature request so that we can address the request separately.

    Thanks,
    Aneita

Comments (27)

  1. Aneita Yang staff
    • changed status to open

    Thanks for the suggestion. I'll open this issue to gauge the interest of other users on this feature. However, the team are currently working on higher priority features, so this is not something that we'll support the near future.

  2. Aneita Yang staff

    Hi everyone,

    Thanks for your interest in this feature.

    We're in the process of investigating this piece of work and I'm interested in chatting with some customers about their needs for rerunning steps. If you would like to discuss your use case / requirements for this feature and how you envision this feature to work, please reach out to me at ayang@atlassian.com and we can schedule a quick chat.

    Aneita

  3. Shane Fast

    Excited for this - How do you think you will handle artifacts for reruns of intermediate steps? reuse the artifacts from the previous run or use a clean slate?

  4. Jason Turner

    Hey, we would like this one as well. The use case that we have is that a step fails due to another dependency outside of the pipelines control. Once that is fixed the step should just re-try.

    I would assume using the previous artifacts, i.e. we generally have buid -> deploy Stage -> deploy Prod. The artifacts need to be consistent between Stage/prod releases.

    Hope that helps

  5. Aidan Hallett

    Our pipeline includes 3 main stages; unit testing, build/deployment, and end to end tests. We want to be able to run the end to end tests without having to run the rebuild and redeploy everything. Being able to rerun specific stages would resolve this issue for us.

  6. Swarup Donepudi

    We use code deploy to deploy our applications after building and running a commit. At times Code Deploy fails with out a reason. Ability to rerun just the deployment step would fix this problem for us.

  7. Rachael Ludwick

    One really frustrating and critical issue here is that without step re-run, if we want "green" builds (e.g. on a PR branch before allowing merge), we are using basically double build minutes whenever there's a transietn out of our control failure on any step. For example, one of our pipelines has a static analysis step that runs a few quick tools that analyze the code/config (e.g. a bundler audit for a rails project). I just had a failure that seems to be a transient Pipelines issue:

    + GIT_LFS_SKIP_SMUDGE=1 git clone --branch="..." ...
    Cloning into '/opt/atlassian/pipelines/agent/build'...
    error: RPC failed; HTTP 502 curl 22 The requested URL returned error: 502 Bad Gateway
    fatal: The remote end hung up unexpectedly
    

    This is during the "Build setup". If we turned on "require passing build" on this repository (which we've considered), I would have to rerun all the steps even though they all passed and this one is transient.

  8. Zee Zali

    The amount of time one loses from rerunning a pipeline because of failed step is insane 😑

    The fact that you have to rerun the entire pipeline from the beginning is so costly.

    Allowing people to rerun individual steps would be absolutely amazing.

  9. Otto Kruse

    Yes re-running a single step would be great (was surprised at first this was not possible yet). So annoying to have to rerun the entire pipeline, because of some step running into an error due to a transient external condition.

  10. Aneita Yang staff

    Hi everyone,

    Good news - the team is currently working on implementing the ability to rerun failed steps and we hope to have something available to users by mid-August. Please continue to watch this ticket for updates.

    Thanks,
    Aneita

  11. Aneita Yang staff

    Hi everyone,

    I'm excited to let you know that the ability to rerun failed steps is now in Pipelines Alpha! For anyone who is already part of the alpha group, the feature is available to you right now. If you're not part of the alpha group but would like to trial the feature, you can sign up to be an alpha customer.

    With this feature, you can now choose whether you want to rerun failed steps or the entire pipeline from the Pipelines UI.

    rerun_steps.png

    For more information about step rerun, check out our documentation.

    If you have any feedback on the feature, please let us know by emailing us at pipelines-feedback@atlassian.com.

    Thanks!
    Aneita

  12. Aneita Yang staff

    Hi everyone,

    We're super excited to announce that the ability to rerun failed steps is now available to all Pipelines users. This lets you save both time and build minutes. For more information about step rerun, check out our documentation.

    If you are interested in the ability to rerun successful steps for the purpose of rolling back changes, I encourage you to vote for / watch issue #16604 which is an open issue to track that functionality separately. If you have any other suggestions or requests that weren't covered as part of this feature, please raise a new feature request so that we can address the request separately.

    Thanks,
    Aneita

  13. Log in to comment