You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@polaris.apache.org by "youngrae81 (via GitHub)" <gi...@apache.org> on 2026/09/16 04:29:11 UTC
Re: [PR] fix: clarify the error when X-Iceberg-Access-Delegation resolves to unsupported remote signing [polaris]
youngrae81 commented on code in PR #5473:
URL: https://github.com/apache/polaris/pull/5473#discussion_r4022493523
##########
CHANGELOG.md:
##########
@@ -110,7 +110,11 @@ request adding CHANGELOG notes for breaking (!) changes and possibly other secti
- Deleting a semantic model now returns HTTP 404 instead of HTTP 500 when the model or its
catalog path disappears after resolution and before the deletion is persisted.
- Return HTTP 404 instead of 500 when a policy or its catalog path disappears after resolution and before deletion.
-
+- Iceberg REST: a request whose `X-Iceberg-Access-Delegation` resolves to remote signing (not yet
+ implemented) now returns the table without delegated access instead of `400 Unsupported access
+ delegation mode: REMOTE_SIGNING`. This applies to `remote-signing` requested alone and to
+ `vended-credentials,remote-signing` against a catalog that cannot vend credentials. A response without
Review Comment:
Done in the latest commit, thanks both. Both `remote-signing` alone and `vended-credentials,remote-signing` against a non-vending catalog now fail with "This catalog cannot vend credentials or sign requests; request without X-Iceberg-Access-Delegation and configure storage credentials on the client" (with "and", as suggested). No behavior change otherwise: the fallback branch, the debug log and the defensive refresh-endpoint tweak are gone, and the docs paragraph is back to its original wording.
One implementation note: I raise the error in the handler's `resolveAccessDelegationModes` rather than inside `DefaultAccessDelegationModeResolver`, since that is where "remote signing is not implemented" already lives and it keeps the resolver's existing tests untouched. Happy to move it if you prefer it in the resolver.
I'll open a follow-up issue for the spec-alignment discussion (server may supply none of the requested mechanisms) as @flyrain suggested. PR title and description updated to reflect the narrower scope.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@polaris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org