1472626 - Error: [Exception... "Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIFaviconService....

Status: RESOLVED FIXED

(Firefox :: Tabbed Browser, defect, P2)

Description

[Mark Banner (:standard8)]

This bug was automatically filed from Sentry: https://sentry.prod.mozaws.net/operations/nightly-js-errors/issues/4426746/

Error: [Exception... "Component returned failure code: 0x80004005 (NS_ERROR_FAILURE) [nsIFaviconService.replaceFaviconDataFromDataURL]"  nsresult: "0x80004005 (NS_ERROR_FAILURE)"  location: "JS frame :: resource:///modules/PlacesUIUtils.jsm :: loadFavicon :: line 192"  data: no]
    at loadFavicon(resource:///modules/PlacesUIUtils.jsm:192:7)
    at loadFavicon(resource:///modules/PlacesUIUtils.jsm:324:5)
    at setIconFromLink(chrome://browser/content/browser.js:3769:7)
    at receiveMessage(chrome://browser/content/browser.js:3743:9)

This is new as of a couple of days ago, blame points to bug 1453751.

Comment 1

[Dave Townsend [:mossop]]

We're catching this and reporting it so I guess it is just expected here. Not sure what else to do here or why it is showing up in sentry.

Comment 2

[Dão Gottwald [:dao]]

Why are we reporting this if it's an expected exception?

Comment 3

[Marco Bonardo [:mak]]

Hm, what's failing exactly in replaceFaviconDataFromDataURL? the API is supposed to complain if it gets invalid input, otherwise we'd hide bugs. It's possible there's just a missing check in loadFavicon and we should not invoke replaceFaviconDataFromDataURL with some invalid input.

Comment 4

[Marco Bonardo [:mak]]

And yes, if it's really expected we should probably analyze the exception and filter out the expected ones.

Comment 5

[Osmose [:osmose, :mkelly]]

(In reply to Dave Townsend [:mossop] from comment #1)
> We're catching this and reporting it so I guess it is just expected here.
> Not sure what else to do here or why it is showing up in sentry.

If it's expected and you want to log it just as a note without collecting, log it as a warning and it won't be collected[1].

[1] https://searchfox.org/mozilla-central/source/browser/modules/BrowserErrorReporter.jsm#249

Comment 6

[Dave Townsend [:mossop]]

(In reply to Marco Bonardo [::mak] from comment #3)
> Hm, what's failing exactly in replaceFaviconDataFromDataURL? the API is
> supposed to complain if it gets invalid input, otherwise we'd hide bugs.
> It's possible there's just a missing check in loadFavicon and we should not
> invoke replaceFaviconDataFromDataURL with some invalid input.

Difficult to say what is failing but we could add more information to the error log to see that in sentry. Looking at the C++ code it looks like we do various checks on the size of the decoded image that we can only really do in C++. We could also change the C++ result in that case to be something more descriptive, maybe NS_ERROR_FILE_TOO_BIG.

Comment 7

[Jeronimo Torti (Please send NI request to Brindusa Tot)]

Hi,
I've encountered this behaviour on Windows 10 with Firefox Nightly 74.0a1 (2020-01-13) (64-bit) in the following page http://webglsamples.org/aquarium/aquarium.html

Regards.

Comment 8

[Augusto Pace]

Hi,

This issue was reproducible on Mac OS 10.14 with Firefox Nightly version 75.0a1 (2020-02-18) . Marked as affected.

Comment 9

[Seba S.]

Hi,

Could reproduce this issue on 77.0a1, Win10 64x.

Regards,

Comment 10

[BugBot [:suhaib / :marco/ :calixte]]

Change the status for beta to have the same as nightly and release.
For more information, please visit auto_nag documentation.

Comment 11

[Mark Banner (:standard8)]

Hi, thank you but there is no need to keep re-verifying this for releases. If it goes away that might be interesting, but we don't need comments to say this is still occurring.

Comment 13

[Masatoshi Kimura [:emk]]

At least we don't have to report 404 Not Found IMO. Is is completely normal that a website does not put a favicon.

Comment 14

[Masatoshi Kimura [:emk]]

Attachment: Bug 1472626 - Stop spamming favicon loading errors with HTTP failures to Browser Console. r?mossop

We already logs all HTTP messages. There is not point in spamming duplicate information.

Comment 15

[Andreea Pavel [:apavel]]

https://hg.mozilla.org/mozilla-central/rev/21ede34c6fc2