7761396303
We occasionally see the this task fail for the first element in the
zuul.projects list with a MODULE FAILURE and a return code of -13
(SIGPIPE) [1]. So far we couldn't identify the root cause, so try to
mitigate this issue by retrying on failure. This solution is similar to
the one used for the "Synchronize repos" task[2].
There is a bug report in Ansible that fits
Since it's only the first element in the loop that is failing while
subsequent elements are successful, we currently have two assumptions:
1. As the task before is using a `delegate_to: localhost' [3],
there might be a problem with Ansible when switching the connection
from localhost to the remote host (node).
2. Since the task before is using the same SSH connection [4] that is
used by Ansible to push the git repository, there might be some
"leftovers" on the connection that make the next task fail.
3. There is also a bug report in Ansible [5] which might be causing
that error.
[1]:
{
"ansible_loop_var": "zj_project",
"changed": false,
"failed": true,
"module_stderr": "",
"module_stdout": "",
"msg": "MODULE FAILURE\nSee stdout/stderr for the exact error",
"rc": -13,
"zj_project": {...}
}
[2]: 3b3495e255/roles/mirror-workspace-git-repos/tasks/main.yaml (L32)
[3]: 3b3495e255/roles/mirror-workspace-git-repos/tasks/main.yaml (L25)
[4]: 3b3495e255/roles/mirror-workspace-git-repos/tasks/main.yaml (L16)
[5]: https://github.com/ansible/ansible/issues/81777
Change-Id: I0c4cb87bb076b9b40c9c446dbe5db437daff5897
71 lines
2.9 KiB
YAML
71 lines
2.9 KiB
YAML
- name: Allow pushing to non-bare repo
|
|
git_config:
|
|
name: receive.denyCurrentBranch
|
|
value: ignore
|
|
scope: local
|
|
repo: "{{ zuul_workspace_root }}/{{ zj_project.value.src_dir }}"
|
|
with_dict: "{{ zuul.projects }}"
|
|
loop_control:
|
|
loop_var: zj_project
|
|
|
|
- name: Synchronize src repos to workspace directory
|
|
command: |-
|
|
{% if ansible_connection == "kubectl" %}
|
|
git push {% if mirror_workspace_quiet %}--quiet{% endif %} --mirror "ext::kubectl --context {{ zuul.resources[inventory_hostname].context }} -n {{ zuul.resources[inventory_hostname].namespace }} exec -i {{ zuul.resources[inventory_hostname].pod }} -- %S {{ zuul_workspace_root }}/{{ zj_project.value.src_dir }}"
|
|
{% else %}
|
|
git push {% if mirror_workspace_quiet %}--quiet{% endif %} --mirror git+ssh://{{ ansible_user }}@{{ ansible_host | ipwrap }}:{{ ansible_port }}/{{ zuul_workspace_root }}/{{ zj_project.value.src_dir }}
|
|
{% endif %}
|
|
args:
|
|
chdir: "{{ zuul.executor.work_root }}/{{ zj_project.value.src_dir }}"
|
|
environment:
|
|
GIT_ALLOW_PROTOCOL: ext:ssh
|
|
with_dict: "{{ zuul.projects }}"
|
|
loop_control:
|
|
loop_var: zj_project
|
|
delegate_to: localhost
|
|
# We occasionally see git pushes in the middle of this loop fail then
|
|
# subsequent pushes for other repos succeed. The entire loop ends up
|
|
# failing because one of the pushes failed. Mitigate this by retrying
|
|
# on failure.
|
|
register: git_push
|
|
until: git_push is success
|
|
retries: 3
|
|
# ANSIBLE0006: Skip linting since it triggers on the "git" command,
|
|
# but push is not supported by ansible git module.
|
|
tags:
|
|
- skip_ansible_lint
|
|
|
|
# Do this as a multi-line shell so that we can do the loop once
|
|
- name: Update remote repository state correctly
|
|
shell: |
|
|
set -eu
|
|
|
|
# Reset is needed because we pushed to a non-bare repo
|
|
git reset --hard
|
|
# Clean is needed because we pushed to a non-bare repo
|
|
git clean -xdf
|
|
# Undo the config setting we did above
|
|
git config --local --unset receive.denyCurrentBranch
|
|
# checkout the branch matching the branch set up by the executor
|
|
git checkout {% if mirror_workspace_quiet %}--quiet{% endif %} {{ zj_project.value.checkout }}
|
|
# put out a status line with the current HEAD
|
|
echo "{{ zj_project.value.canonical_name }} checked out to:"
|
|
git log --pretty=oneline -1
|
|
args:
|
|
chdir: "{{ zuul_workspace_root }}/{{ zj_project.value.src_dir }}"
|
|
with_dict: "{{ zuul.projects }}"
|
|
loop_control:
|
|
loop_var: zj_project
|
|
# We occasionally see the this task fail for the first element in the
|
|
# zuul.projects list with a MODULE FAILURE and a return code of -13
|
|
# (SIGPIPE). This may be caused by
|
|
# https://github.com/ansible/ansible/issues/81777
|
|
# Try to mitigate this issue by retrying on failure.
|
|
register: git_update
|
|
until: git_update is success
|
|
retries: 3
|
|
# ANSIBLE0006: Skip linting since it triggers on the "git" command,
|
|
# but we prefer the shell above
|
|
tags:
|
|
- skip_ansible_lint
|