pkg/payload: Initial work on update -> reconciliation wording#194
pkg/payload: Initial work on update -> reconciliation wording#194wking wants to merge 6 commits intoopenshift:masterfrom
Conversation
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: wking 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 |
openshift-ci-robot
added
approved
|
CC @abhinavdahiya, @smarterclayton Not sure if there's anything I can do to help with review, this is a fair bit bigger than #185, and that's been pretty quiet ;). |
wking
force-pushed
the
update-payload-to-release-image
branch
from
86704fa to
13630b3
Compare
openshift-ci-robot
removed
the
needs-rebase
|
I think the CVO reacts to Updates and not Payloads. So I don't completely agree with the move. like payload.UpdateError - > payload.Error , personally i think former is more clearer. haven't looked at the change in depth |
The CVO has three modes, all of which run reconciling sync loops, and only one of which is an update.
But we are generating these errors outside of updates, as seen in rhbz#1711964. These issues are payload-specific (which is already covered by the package scoping), but they aren't all upgrade-specific.
Can you put your finger on why? Getting a short text rendering of an instance of the payload package's error type seems like a good fit for "error type method" to me (vs. "grab a separate function from the payload package and pass it portions of the error type"). |
wking
force-pushed
the
update-payload-to-release-image
branch
from
13630b3 to
27ee47d
Compare
Narrow pivot to start moving the focus away from upgrades and towards reconciliation. This is the main type out of this package, so just echo the package name.
Narrow pivot moving the focus away from upgrades and towards reconciliation. This loader returns the main type out of this package, so drop the "what is being loaded" qualifier from the function name to avoid stuttering the package name.
This is the only error type in this package, so there's no need to be more specific. This also moving the focus away from upgrades, to avoid marginalizing reconciliation.
There's no need for a separate function here, and using a method gives us access to additional Error properties if we want them (although I haven't pivoted to take advantage of that yet).
…stsError) The errors feeding this have little to do with integrity. And they aren't upgrade-specific; disk corruption or similar during regular reconciliation could also lead to this error.
This allows us to avoid having two separate but very similar blocks of code for converting these errors to succint, admin-oriented descriptions.
wking
force-pushed
the
update-payload-to-release-image
branch
from
27ee47d to
5242487
Compare
|
We should definitely not use the word payload anywhere in messages or user interface - update or version are acceptable, but payload is not part of our public lexicon. It’s fine internally |
| return e.Nested | ||
| } | ||
|
|
||
| func (e *Error) Summary() string { |
There was a problem hiding this comment.
This was separate because it was a specific domain error being applied - the error message for API short status is not the same as the message in other contexts. Summary isn’t a property of the error, it is an interpretation of it.
So would prefer not to do this part.
|
@wking: The following tests failed, say
Full PR test history. Your PR dashboard. DetailsInstructions 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/test-infra repository. I understand the commands that are listed here. |
|
@wking: The following tests failed, say
Full PR test history. Your PR dashboard. DetailsInstructions 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/test-infra repository. I understand the commands that are listed here. |
|
Issues go stale after 90d of inactivity. Mark the issue as fresh by commenting If this issue is safe to close now please do so with /lifecycle stale |
openshift-ci-robot
added
the
lifecycle/stale
|
Stale issues rot after 30d of inactivity. Mark the issue as fresh by commenting If this issue is safe to close now please do so with /lifecycle rotten |
openshift-ci-robot
removed
the
lifecycle/stale
openshift-ci-robot
added
the
lifecycle/rotten
|
Rotten issues close after 30d of inactivity. Reopen the issue by commenting /close |
openshift-ci
Bot
added
the
needs-rebase
|
@wking: PR needs rebase. DetailsInstructions 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/test-infra repository. |
|
@openshift-bot: Closed this PR. 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/test-infra repository. |
6 participants
A bunch of small pivots to start moving the package API and error messages away from "update" phrasing and towards generic reconiciliation phrasing. When complete, this transition will reduce confusion like rhbz#1711964.