-
Notifications
You must be signed in to change notification settings - Fork 49
Conversation
closes: #1039 Signed-off-by: knrt10 <kautilya@kinvolk.io> Co-authored-by: Mateusz Gozdek <mateusz@kinvolk.io>
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.
Please write an e2e test to verify this behaviour.
This might be complex, as our e2e test suite do not have access to cluster configuration to modify it... What is your idea to test this @surajssd ? |
@invidian I had a talk with Suraj offline and I think this should be like this.
What do you think? I am little confused. Can we update it via client-go? |
It can only be updated via Helm render, as it's Helm template which generates the checksum. |
In that case we can have a render test locally, which verifies that the annotations change on each render. |
cfd9b44
to
c2952ea
Compare
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.
Looks OK, though I think Fatalf
should be used instead of Errorf
, as most of the errors will cause further code to panic.
On every render helm template will generate the checksum which will be different for unique config. Which in turn will regenerate pods for deployment on config change. Signed-off-by: knrt10 <kautilya@kinvolk.io>
c2952ea
to
405c204
Compare
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.
Looks OK
closes: #1039
Signed-off-by: knrt10 kautilya@kinvolk.io
Co-authored-by: Mateusz Gozdek mateusz@kinvolk.io