861284 - Error reporting for opening (signed) packaged apps is not good enough for QA to diagnose problems

Status: RESOLVED WONTFIX

(Core Graveyard :: DOM: Apps, defect)

Description

[Brian Smith (:briansmith, :bsmith, use NEEDINFO?)]

nsIX509CertDB.openSignedJarFileAsync reports these error messages, all of which get rolled up into "INVALID_SIGNATURE." This makes it impossible for QA to diagnose problems without stepping through the code in the debugger.

We should report each error distinctly, at least to the error console (so QA can use logcat to see the more specific error). If we report the distinct errors to Gaia then there should be a Gaia bug for handling these new error codes by showing distinct error messages.

The data below is copy/pasted/modified from js/xpconnect/src/xpc.msg:

NS_ERROR_SIGNED_JAR_NOT_SIGNED:
"The JAR is not signed."

NS_ERROR_SIGNED_JAR_MODIFIED_ENTRY:
"An entry in the JAR has been modified after the JAR was signed."

NS_ERROR_SIGNED_JAR_UNSIGNED_ENTRY,
"An entry in the JAR has not been signed."

NS_ERROR_SIGNED_JAR_ENTRY_MISSING
"An entry is missing from the JAR file."

NS_ERROR_SIGNED_JAR_WRONG_SIGNATURE
"The JAR's signature is wrong."

NS_ERROR_SIGNED_JAR_ENTRY_TOO_LARGE
"An entry in the JAR is too large."

NS_ERROR_SIGNED_JAR_ENTRY_INVALID
"An entry in the JAR is invalid."

NS_ERROR_SIGNED_JAR_MANIFEST_INVALID
"The JAR's manifest or signature file is invalid."

Comment 1

[Brian Smith (:briansmith, :bsmith, use NEEDINFO?)]

(In reply to Brian Smith (:bsmith) from comment #0)
> nsIX509CertDB.openSignedJarFileAsync reports these error messages, all of
> which get rolled up into "INVALID_SIGNATURE." This makes it impossible for
> QA to diagnose problems without stepping through the code in the debugger.

Forgot to say that this is also important so that the marketplace and server ops teams can diagnose potential problems in their side (the actual signing of the apps).

Comment 2

[Jason Smith [:jsmith]]

I take it this is related to bug 841569, right?

Comment 3

[Brian Smith (:briansmith, :bsmith, use NEEDINFO?)]

(In reply to Jason Smith [:jsmith] from comment #2)
> I take it this is related to bug 841569, right?

Yes. This was spun out of my conversation with PMPCat about that bug and about how we can communicate and resolve issues faster.

Comment 4

[Jason Smith [:jsmith]]

Not a blocker, but we might this to improve debugging workflow for signature-based bugs, given that we ran into a whole lot trouble in debugging speed on bug 841569.

tracking? for consideration

Comment 5

[Sylvestre Ledru [:Sylvestre]]

Core Graveyard / DOM: Apps is inactive. Closing all bugs in this component.