Basic Checkout of build fails on 22.04, works on 20.04

[smanleyjr]

I did a basic run checkout step using cimg/base:2022.06-22.04 and cimg/base:2022.06-20.04.... Why is 22.04 failing? but 20.04 works?

Error Seen when failing

Using SSH Config Dir '/home/circleci/.ssh' git version 2.36.1 Cloning git repository Cloning into '.'... error: cannot create async thread: Operation not permitted fatal: fetch-pack: unable to fork off sideband demultiplexer

exit status 128

Simple config

# Use the latest 2.1 version of CircleCI pipeline process engine. See: https://circleci.com/docs/2.0/configuration-reference
version: 2.1
# Use a package of configuration called an orb.
orbs:
  # Declare a dependency on the welcome-orb
  welcome: circleci/[email protected]
jobs:
  build:
    working_directory: /home/circleci/project
    docker:
     # - image: cimg/node:12.20.2
      - image: cimg/base:2022.06-22.04
     # - image: cimg/base:2022.06-20.04

    steps:
      - checkout
      - run: echo $(ls -a) 
      
workflows:
  version: 2
  test:
    jobs:
        - build:
            context: 
              - smanley1-XXX17-Sndbx

[FelicianoTech]

Hey @smanleyjr , is this still an issue?

There was an issue with newer versions of git that cropped up when git fixed a CVE and thus changed its behavior. We ended up fixing that. Might this be the same issue and now it's fixed?

[brianrourkeboll]

I am seeing this issue pop up with cimg/base:current (= cimg/base:2022.09). It does not happen with cimg/base:2022.08.

For me, this is happening on a CircleCI Server installation. I see some evidence elsewhere on the 'net that this might require that the Docker engine we're using be updated.

[bisswanger]

We are seeing the same since 2 days on the cimg/go:1.18 and cimg/go:1.19 It's fine for cimg/go:1.19.1 (which carries git 2.37 still)

[FelicianoTech]

Does anyone have a public project they can link to that shows the issue coming up? If you not, can someone please open a support ticket with their private build URL and link back to this GitHub issue? I need to be able to reproduce this issue in order to help and so far I haven't been able.

I see some evidence elsewhere on the 'net that this might require that the Docker engine we're using be updated.

I've seen this too with people using setup_remote_docker. A newer version should be set for that.

[bisswanger]

@felicianotech : unfortunately I only do see it an an instance internally at the company (should be self-hosted). Also few colleagues internally confirmed the same issue.

Could indeed be the Docker version. What I tested:

• Pull the go 1.19 image locally on my Mac
• Launch bash on the image
• Execute the same commands as CircleCI is doing

With that the git clone command completes successfully, i.e. no crash.

[denysvitali]

Hey @bisswanger: do you use VMware Tanzu internally? We're facing this issue at @swisscom and @swisspost using VMware Tanzu.

See:

• https://github.com/argoproj/argo-cd/issues/9888#issuecomment-1271819234
• https://github.com/argoproj/argo-cd/issues/9809#issuecomment-1269543055

[FelicianoTech]

Previous work seems to suggest that this issue is related to users not using CircleCI Cloud. Instead, CircleCI Server seems to be involved. In that scenario, an older version of Docker is being used to power the Docker executor.

Is anyone here having this issue and doesn't fit this description?