You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
*: Port from Update to Release for ClusterVersion status
The next/last Actual comparison avoids:
invalid operation: next.Actual == last.Actual (struct containing []string cannot be compared)
The image pullspec is sufficient for that matching. The test suite's
expected LastProgress changes are because the shift from:
{Version: "4.0.1", Image: "image/image:1"}
to:
{Version: "1.0.0-abc", Image: "image/image:1"}
no longer counts as a LastProgress-bumping change (because we're now
only comparing the Image pullspecs).
Also adjusts the upstream/Cincinnati retrieval and payload loading
logic to preserve the additional metadata that we're now exposing.
Storing a Release on the Operator, instead of extending the previous
releaseVersion/releaseImage pattern, makes it easy to keep track of
payload-loaded metadata, so we can fall-back to upstream metadata for
properties where the payload has no opinion, as described in the
enhancement.
In findUpdateFromConfigVersion, if some crazy upstream service
declared a node with a matching version but no image, we used to say
"sorry, no match" without looking through history. Now we skip the
useless entry and move on to consider later entries and history. We
only say "no match" if we have no image pullspec associated with the
requested version available.
Copy file name to clipboardExpand all lines: docs/user/status.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -110,7 +110,7 @@ It could also be an overloaded or otherwise failing Cincinnati server.
110
110
111
111
### ResponseInvalid
112
112
113
-
The Cincinnati server returned a response that was not valid JSON.
113
+
The Cincinnati server returned a response that was not valid JSON or is otherwise corrupted.
114
114
115
115
This could be caused by a buggy Cincinnati server.
116
116
It could also be caused by response corruption, e.g. if the configured `upstream` was in the clear over HTTP or via a man-in-the-middle HTTPS proxy, and an intervening component altered the response in flight.
0 commit comments