Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Jul 26 2015 16:49
    Alex Suraci edited As a user of Fly I would like to be able to hijack a particular one-off build feature to Concourse - view »
  • Jul 26 2015 16:46
    Alex Suraci moved and scheduled As a user of Fly I would like to be able to hijack a particular one-off build story before 'If the beacon can't reach the TSA (or it stops seeing heartbeats) then it should disconnect and keep retrying' in Concourse - view »
  • Jul 26 2015 16:46
    Alex Suraci moved Remove conditionals story before 'CI should run a basic test against concourse/lite virtualbox before publishing it' in Concourse - view »
  • Jul 26 2015 16:46
    Alex Suraci added comment: "This would help make testflight's flying suite safe to run in parallel. Right now its hijack test has a race since it just does fly execute -> fly hijack, and may hijack another test's container." on Concourse - view »
  • Jul 26 2015 00:59
    Alex Suraci accepted Support reading target address from a file feature on Concourse - view »
  • Jul 26 2015 00:59
    Alex Suraci accepted BOSH deployment resource should support filling in thedirector_uuidautomatically feature on Concourse - view »
  • Jul 26 2015 00:22
    Alex Suraci moved Support reading target address from a file story before 'BOSH deployment resource should support filling in the director_uuid automatically' in Concourse - view »
  • Jul 26 2015 00:22
    Alex Suraci moved BOSH deployment resource should support filling in thedirector_uuidautomatically story before 'v0.59.0' in Concourse - view »
  • Jul 26 2015 00:22
    Alex Suraci delivered Support reading target address from a file feature to Concourse - view »
  • Jul 26 2015 00:22
    Alex Suraci delivered BOSH deployment resource should support filling in thedirector_uuidautomatically feature to Concourse - view »
  • Jul 26 2015 00:22
    Alex Suraci finished Support reading target address from a file feature in Concourse - view »
  • Jul 26 2015 00:22
    Alex Suraci finished BOSH deployment resource should support filling in thedirector_uuidautomatically feature in Concourse - view »
  • Jul 25 2015 23:24
    Alex Suraci started Support reading target address from a file feature on Concourse - view »
  • Jul 25 2015 23:24
    Alex Suraci estimated Support reading target address from a file feature as 1 point in Concourse - view »
  • Jul 25 2015 23:19
    Alex Suraci started BOSH deployment resource should support filling in thedirector_uuidautomatically feature on Concourse - view »
  • Jul 25 2015 23:19
    Alex Suraci estimated BOSH deployment resource should support filling in thedirector_uuidautomatically feature as 0 points in Concourse - view »
  • Jul 25 2015 23:18
    Alex Suraci added BOSH deployment resource should support filling in thedirector_uuidautomatically feature to Concourse - view »
  • Jul 25 2015 23:14
    Alex Suraci accepted bosh-deployment resource shouldn't require specifying full https://...:25555 uri feature on Concourse - view »
  • Jul 25 2015 22:30
    Alex Suraci moved bosh-deployment resource shouldn't require specifying full https://...:25555 uri story before 'v0.59.0' in Concourse - view »
  • Jul 25 2015 22:30
    Alex Suraci delivered bosh-deployment resource shouldn't require specifying full https://...:25555 uri feature to Concourse - view »
Regev Brody
@regevbr
Is there a way to pass key values between jobs instead of using resources? It seems like an overkill to write to s3/git some key values just to pass them along the pipeline...
xlr-8
@xlr-8
Not AFAIK
Regev Brody
@regevbr
Bummer... Ill open an issue I think this is only natural to support it
Regev Brody
@regevbr
concourse/concourse#1804
xlr-8
@xlr-8
This message was deleted
Regev Brody
@regevbr
Hi
Anyone here?
I have a job that lock a remote resource for its own use
I wrote an on_failure step that will release that lock in case of job failure
The issue is I have not founf a way to pass the relevant lock data between the plans so the on_failure will know what to unlock
Zenobius Jiricek
@airtonix
@regevbr it's almost as if you need to spin up a mongo container and create a mongo-resource to get/set values
Zenobius Jiricek
@airtonix

I'm using a task that commits to a git locally cloned repo. sometimes (randomly) the git clone ./here ./there results in a detached head state thus an exit code 1 and then the whole concourse job fails.

the task is :

  - task: commit to git
    config:
      platform: linux

      image_resource:
        type: docker-image
        source: {repository: concourse/git-resource}

      inputs:
      - name: repo
      - name: docker-compose
      - name: docker-compose-build
      outputs:
      - name: repo-modified

      run:
        path: sh
        args:
        - -exc
        - |
          git config --global user.name "concourse"
          git config --global user.email "concourse@no.reply"

          git clone repo repo-modified
          cd repo-modified
          git pull
          cp ../docker-compose/* ../docker-compose-build/* ./
          git add ./docker-compose.yml ./docker-compose--build.yml
          if ! git diff-index --quiet HEAD --; then
            git commit -m "CI: update docker-compose manifests with new version [skip ci]"
          else
            echo "No changes to commit. skipping"
          fi
tried using git clone --local --no-hardlinks ./repo ./repo-modified but still arrives at detached HEAD
Tom Beute
@beuteiful_twitter
anyone have an idea how to make a concourse resource that can ignore its own put changes?
Tom Beute
@beuteiful_twitter
anyone here can help with the git resource?
what im seeing happening is when it pushes to github it actually detects its own changes and triggers the build again
Zenobius Jiricek
@airtonix
I think I figured it out. You need to have the same repo configured as a resource for a trigger and again under another name for the put operation
Tom Beute
@beuteiful_twitter
anyone here?
Khaled Hussain
@codehopper-uk
Anyone have issues with Teams and OAuth2 and Google in particular. I have a weird issue with the redirect url in the request to google having a period in it and Google seems therefore unable to match it against the configured/allowed redirect url in Google account settings.
sky-eduardolopes
@sky-eduardolopes
Hi guys, I have a Linux worker that I only want to use for Android tests. How can I make sure it's not used for the rest of the pipeline?
Matt Narrell
@mnarrell
i see notation like this in the tutorial FLY_TEAM: (( grab meta.team ))
what is this grab directive? where are things like that documented?
Lukas M
@lukasmrtvy
Can anyone give me example of post commit hook from git to trigger concourse job?
Steven Gonsalvez
@stevengonsalvez
what is this grab directive? where are things like that documented? +1
Kevin Fries
@kevin-fries
does anyone have a good example site that could help me set up a server that can call packer, ansible, and terraform from make files?
also looking for a good example of a trigger by either git checkin or always at a scheduled day and time (i.e. Sundays at 4am)
I would also love if I could see something that can trigger based on AMI being updated
John Torres
@johnt337
Hey all
I have two questions that I was hoping for some direction around
  1. Any good suggestions for handling approval processes that exist outside of the pipeline. Like a service-now type tool. Or a manual approval process?
  2. I’m using AWS roles for permission to resources, how best to handle AWS security/session tokens that expire and have to be updated frequently
This message was deleted
Kris Krishna
@kriskrishna
how to pass google cloud credentials in Concourse pipeline
Pavel Prisyazhnyuk
@templar15.2012_gitlab
Hi all! Does concourse has anything like this https://docs.gitlab.com/runner/configuration/runner_autoscale_aws/ ?
Mohammad Al mouallem
@mhd999
Hi!
git-resource: Is there away to push a feature branch using put step?
Mohammad Al mouallem
@mhd999

What am trying to achieve is creating a branch in a job then push to the resource repo to open a pull request later. :point_up_2:

Thanks.

Lukas M
@lukasmrtvy
Anyone using Traefik and Concourse?
Jorge Viana
@jorgeviana
Hello, I was wondering if there is a way to test Concourse pipelines code. Does anybody knows how this could be done?
fmcejudo
@fmcejudo
Hi, I am trying to create a pipeline in which in one of the task it pushes a docker image to AWS ECR, but I have to be login with aws-cli ... so, are you aware if there any resource_type or docker image which let me do so
?
Jacob Martin
@jacob-hd
@fmcejudo This channel is dead
fmcejudo
@fmcejudo
do we have any alternative?
btw, @jacob-hd thanks for taking care of missing people
Jeff Jenner
@thetooth
can someone help me with the basics of getting artifacts into a put task?
i've tried using an output, copying dockerfile and binaries into this folder, but put always complains it can't find the dockerfile
domdom8
@domdom8
i wonder is there a room for spinnaker ?
Alex Holyoake
@aholyoake-bc
Hi Guys, we're currently having a major outage after rebuilding our worker nodes - Concourse is showing run check step: run check step: find or choose worker for container: no workers satisfying: resource type 'git' for all resource types.
However, fly -t main workers -d indicates that all the workers are present and in the running state, and have all the resource types that you'd normally expect. Including git, docker-image and registry-image
any suggestions for even starting to debug this would be very appreciated
firewall rules all appear to be correctly configured (not that these have been changed since the cluster was working). We've also tried upgrading to 6.5.1 to see if that would help, but unfortunately it hasn't.
Alex Holyoake
@aholyoake-bc
We just found the issue - we had a CONCOURSE_TAG variable set which was empty. I don't know why, but this stopped all the workers from functioning
I'll file a bug for this behaviour when the tag is empty. Sorry for the noise!
Alex Suraci
@vito
hi all, sorry for the confusion, this room should have been deleted a long time ago - we use Discord: https://discord.gg/MeRxXKW