Currently, we capture the chef-client output on the target node. However, if the run is successful we throw away that output. There are also failure scenarios where we do not display the output to the user but it would be helpful to do so.
Describe the Need:
We want to capture and store the chef-client output for both successful and unsuccessful runs. Users should be able to configure how they receive this information - if many of their runs are successful, displaying it on STDOUT every time is just noise.
None - we manually configure chef-client to output to stdout via the workstation.rb
Questions
How will users configure this feature?
Where will output be stored?
Do we want to save chef-client output on the target node in addition to piping it back to the workstation?
In the current failure scenario we output STDOUT to the ChefApply::Log - do we want to change that as part of this implementation? Or maybe we always store output into the ChefApply::Log?
Describe the Enhancement:
Currently, we capture the chef-client output on the target node. However, if the run is successful we throw away that output. There are also failure scenarios where we do not display the output to the user but it would be helpful to do so.
Describe the Need:
We want to capture and store the chef-client output for both successful and unsuccessful runs. Users should be able to configure how they receive this information - if many of their runs are successful, displaying it on STDOUT every time is just noise.
We also need to ensure that the chef-client output can be viewed in all scenarios (EG, https://github.com/chef/chef-apply/issues/133).
Current Alternative
None - we manually configure chef-client to output to stdout via the
workstation.rbQuestions
ChefApply::Log- do we want to change that as part of this implementation? Or maybe we always store output into theChefApply::Log?