-
Notifications
You must be signed in to change notification settings - Fork 19
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merge updates from ocm-plus repo #16
Merge updates from ocm-plus repo #16
Conversation
* lease api updates Signed-off-by: ckandag <ckandaga@redhat.com> * fix for kind test Signed-off-by: ckandag <ckandaga@redhat.com> * fix quote Signed-off-by: ckandag <ckandaga@redhat.com> * debug Signed-off-by: ckandag <ckandaga@redhat.com> * fix hubconfig logic Signed-off-by: ckandag <ckandaga@redhat.com> * fix fmt Signed-off-by: ckandag <ckandaga@redhat.com>
…-cluster-management-io#133) * change the way compliance is handled across multiple namespaces Signed-off-by: Gus Parvin <gparvin@redhat.com> * organize changes more from review comments Signed-off-by: Gus Parvin <gparvin@redhat.com> * re-organize checks based on review comments Signed-off-by: Gus Parvin <gparvin@redhat.com>
…nagement-io#135) Signed-off-by: Will Kutler <wkutler@redhat.com>
Signed-off-by: ckandag <ckandaga@redhat.com>
…t-io#140) * workaround to not add extras Signed-off-by: Will Kutler <wkutler@redhat.com> * fix sort failure Signed-off-by: Will Kutler <wkutler@redhat.com> * add fix and test Signed-off-by: Will Kutler <wkutler@redhat.com> * fmt Signed-off-by: Will Kutler <wkutler@redhat.com> * clean up Signed-off-by: Will Kutler <wkutler@redhat.com> * cleanup 2 Signed-off-by: Will Kutler <wkutler@redhat.com> * remove extra compliancetypes Signed-off-by: Will Kutler <willkutler@gmail.com>
Signed-off-by: ckandag <ckandaga@redhat.com>
Signed-off-by: Gus Parvin <gparvin@redhat.com>
* sort template in merge fn Signed-off-by: Will Kutler <wkutler@redhat.com> * fmt Signed-off-by: Will Kutler <wkutler@redhat.com> * test logs Signed-off-by: Will Kutler <wkutler@redhat.com> * logs 2 Signed-off-by: Will Kutler <wkutler@redhat.com> * clean up logs and fix test Signed-off-by: Will Kutler <wkutler@redhat.com> * fix list compare Signed-off-by: Will Kutler <wkutler@redhat.com> * fmt Signed-off-by: Will Kutler <wkutler@redhat.com> * fix mustonlyhave Signed-off-by: Will Kutler <wkutler@redhat.com> * remove logs Signed-off-by: Will Kutler <wkutler@redhat.com> * add test to verify fix Signed-off-by: Will Kutler <wkutler@redhat.com> * rename function Signed-off-by: Will Kutler <wkutler@redhat.com>
* Refactor mergeArrays Adjusted conditionals with equivalent logic, renamed a variable. Signed-off-by: Justin Kulikauskas <jkulikau@redhat.com> * Remove dead code Signed-off-by: Justin Kulikauskas <jkulikau@redhat.com>
* add comments 1/2 Signed-off-by: Will Kutler <wkutler@redhat.com> * add comments 2/2 Signed-off-by: Will Kutler <wkutler@redhat.com> * fmt Signed-off-by: Will Kutler <wkutler@redhat.com> * comment Signed-off-by: Will Kutler <wkutler@redhat.com> * add new block to merge Signed-off-by: Will Kutler <wkutler@redhat.com> * remove unnecessary sort Signed-off-by: Will Kutler <wkutler@redhat.com> * remove unnecessary sort 2 Signed-off-by: Will Kutler <wkutler@redhat.com>
The "templates" package was split out to its own Go module in: stolostron/go-template-utils#2 Signed-off-by: mprahl <mprahl@users.noreply.github.com>
…o#149) * try removing merge block Signed-off-by: Will Kutler <wkutler@redhat.com> * add e2e test Signed-off-by: Will Kutler <wkutler@redhat.com> * fix list handling when len(existing) < len(template) Signed-off-by: Will Kutler <wkutler@redhat.com> * remove redundant mustonlyhave branch Signed-off-by: Will Kutler <wkutler@redhat.com>
…ster-management-io#150) Signed-off-by: Will Kutler <wkutler@redhat.com>
* Add test for whitespace in list Currently, if an item in a list has whitespace at the beginning or end, it won't get matched by the same item because only one of them is trimmed. As a result, and enforce policy with an item like that will keep appending duplicates of the item on the list indefinitely. This test demonstrates the issue with a Deployment. Signed-off-by: Justin Kulikauskas <jkulikau@redhat.com> * Stop trimming some items in lists This should resolve the issue described in the previous commit. Signed-off-by: Justin Kulikauskas <jkulikau@redhat.com>
Signed-off-by: Dale Haiducek <dhaiduce@redhat.com>
Signed-off-by: Will Kutler <wkutler@redhat.com>
This updates the dependency on the go-template-utils module. Resolves stolostron/backlog#14952 Signed-off-by: mprahl <mprahl@users.noreply.github.com>
Signed-off-by: Will Kutler <wkutler@redhat.com>
…nt-io#156) See the release notes for more information: https://github.com/open-cluster-management/go-template-utils/releases/tag/v1.1.0 Signed-off-by: mprahl <mprahl@users.noreply.github.com>
…n-cluster-management-io#159) Signed-off-by: Gus Parvin <gparvin@redhat.com>
…agement-io#160) This fixes the situation where the controller is comparing what is in the policy versus what already exists, and the object being compared contains an integer. The code in the mergeSpecs function marshals and unmarshals the existing object and the object in the policy. Presumably this is to create a copy of the objects before modification. By doing so with the encoding/json package, it converts any integers to float64. This caused an issue in the checkFieldsWithSort function which compared the existing object with the merged object, since the existing object referenced here did not go through the marshal and unmarshal process, but the merged object did. This meant that the integers in the merged object were of type float64, but they remained integers in the existing object. When checkFieldsWithSort converts both integers to strings to compare, if the one of type float64 is a large enough number, it gets printed in scientific notation where as the one that remained of type integer does not. This would cause a mismatch even though it should have matched. The solution is quite simple. Just start using the json package from Kubernetes so that the same unmarshaling process occurs in all cases. Signed-off-by: mprahl <mprahl@users.noreply.github.com>
Signed-off-by: ckandag <ckandaga@redhat.com>
This makes the code consistent with the policy propagator in: stolostron/governance-policy-propagator#101 Signed-off-by: mprahl <mprahl@users.noreply.github.com>
This is to include the fix from: stolostron/go-template-utils#18 Signed-off-by: mprahl <mprahl@users.noreply.github.com>
This fixes a regression as described in: stolostron/go-template-utils#19 Signed-off-by: mprahl <mprahl@users.noreply.github.com>
…ment-io#166) Deep copy the policy before passing it to handleObjectTemplates since even though handleObjectTemplates accepts a copy (i.e. not a pointer) of the policy, policy.Spec.ObjectTemplates is a slice of pointers, so any modifications to the objects in that slice will be reflected in the PolicyMap cache, which can have unintended side effects. This manifested itself when a policy has templates which when resolved, still have intentional template delimiters (i.e. {{ and }}) in the result. On the first time being processed, everything would go well. On the second or third time, when the policy cache was not updated again by the reconciler, the policy in the cache already had resolved templates. This would lead to the templates code trying to reprocess the already resolved policy. Additionally to this, this caused some updates to referenced resource objects in templates to not get properly updated in the policy. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2007575 Signed-off-by: mprahl <mprahl@users.noreply.github.com>
* Annotation to disable templates Signed-off-by: ckandag <ckandaga@redhat.com> * fix glog formatting Signed-off-by: ckandag <ckandaga@redhat.com>
…o#170) * store api resource list in case of failure Signed-off-by: Will Kutler <wkutler@redhat.com> * fmt Signed-off-by: Will Kutler <wkutler@redhat.com> * ignore error if partial list is returned Signed-off-by: Will Kutler <wkutler@redhat.com>
Signed-off-by: Justin Kulikauskas <jkulikau@redhat.com>
/hold t o merge manually |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Wow! Thanks for tackling this! I looked pretty quickly and noted one thing that stuck out. I'm hoping the tests are capturing anything I might have missed.
Signed-off-by: Justin Kulikauskas <jkulikau@redhat.com>
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: dhaiducek, JustinKuli The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
Brings in commits up to the current "release-2.4" branch from the other repo.