rorfs for cacheless lacks write access to /tmp/ to generate cache#3640
Conversation
Signed-off-by: grokspawn <jordan@nimblewidget.com>
Signed-off-by: grokspawn <jordan@nimblewidget.com>
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: perdasilva The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
f631348
into
operator-framework:master
|
/cherry-pick release-4.20 |
|
@perdasilva: cannot checkout DetailsIn response to this:
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. |
|
lol whoops |
Description of the change:
when we are using
extractContent, but w/oextractContent.CacheDir, ensure that we have write permissions to/tmp/by creating an emptydir volume there.Motivation for the change:
When there is no provided
CacheDir, our recent changes to setSecurityContext{ReadOnlyFilesystem: ptr.To(true)}will not allow writing to/tmp/. However,opmwill need to generate a cache and cannot be told another directory to use.Architectural changes:
Testing remarks:
#3614 passed CI with no errors, both unit which covered ExtractContent functionality including w/o CacheDir.
This is because while the unit tests validated the shape of the resulting pods but not the pods' viability in a cluster (i.e. could the pods walk the initcontainer chain successfully and bring the GRPC service up).
ginkgo test "gRPC address catalog source" does evaluate catsrc pod viability using extractContent, but never as cacheless.
Reviewer Checklist
/doc[FLAKE]are truly flaky and have an issue