[
https://issues.apache.org/jira/browse/CAMEL-24535?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Work on CAMEL-24535 started by Andrea Cosentino.
------------------------------------------------
> camel-tensorflow-serving: the Target and Credentials headers are declared but
> never read
> ----------------------------------------------------------------------------------------
>
> Key: CAMEL-24535
> URL: https://issues.apache.org/jira/browse/CAMEL-24535
> Project: Camel
> Issue Type: Bug
> Components: camel-ai
> Reporter: Andrea Cosentino
> Assignee: Andrea Cosentino
> Priority: Major
> Fix For: 4.23.0
>
>
> TensorFlowServingConstants declares TARGET (line 28) and CREDENTIALS (line
> 31) as @Metadata headers, advertised through the headersClass reference on
> the endpoint (line 37). Neither header is ever read anywhere in the module (a
> grep for both constants returns zero reads outside the declaration).
> The gRPC channel is built once in TensorFlowServingProducer.doInit() from the
> endpoint configuration only, so a per-exchange Target/Credentials override
> supplied as a header cannot take effect. Users who set these headers
> expecting to redirect or re-authenticate a single call get silently ignored
> behaviour. The sibling KServe module declares neither header, which confirms
> these two are cruft carried over during scaffolding.
> Fix: remove the two dead constants (a per-exchange target/credentials
> override cannot work against a single init-time channel), or actually honour
> them per exchange by resolving the channel/credentials from the message
> headers when present.
> _Claude Code on behalf of oscerd_
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
