Skip to main content
TheUsualOne
Known Participant
September 20, 2019
Question

Plug-in Failed and Flash Player plug-in crash loading from webview in MacOS Mojave, Catalina Beta 8

  • September 20, 2019
  • 2 replies
  • 1325 views

Are others experiencing a flash player plug-in crash in WebKitPluginHost?

 

Our application uses a WebView in the WebKit framework to load an HTML file which loads an SWF file. When doing this, there is a crash in WebKitPluginHost for the Marcomedia Flash Player Plug-in. I see this in the log:

 

Exception Type:        EXC_BAD_ACCESS (Code Signature Invalid

Termination Reason:    Namespace CODESIGNING, Code 0x2

 

If I turn off System Integrity Protection (SIP) it works.

 

Are others experiencing this problem and if so does Adobe know about it and plan to fix it? It seems like the plug-in does not have proper signing or notarization.

    This topic has been closed for replies.

    2 replies

    steve-awesome_2020
    Participant
    March 9, 2020

    Really?

    TheUsualOne
    Known Participant
    September 23, 2019

    Ran this:

    codesign --verify --deep --verbose "/Library/Internet Plug-Ins/Flash Player.plugin" /Library/Internet Plug-Ins/Flash

    Player.plugin

     

    a sealed resource is missing or invalid In subcomponent: /Library/Internet Plug-Ins/Flash Player.plugin/Contents/PlugIns/FlashPlayer-10.6.plugin

    file added: /Library/Internet Plug-Ins/Flash Player.plugin/Contents/PlugIns/FlashPlayer-10.6.plugin/Contents/Resources/plugin.vch

     

    Removed plugin.vch.

     

    Codesignature was then valid but still wouldn't work. So I don't think the problem is the code signature. Maybe it needs notarized.

    jeromiec83223024
    Inspiring
    September 23, 2019
    We've definitely been doing a lot of notarization work. We shipped a beta build yesterday. Are you seeing this problem there?
    TheUsualOne
    Known Participant
    September 24, 2019
    I tried the Flash Player beta dated Sept. 24, 2019. It still crashes but wtih an SIGABRT excetption, abort() called instead. I think that is progress. I have a bug report opened here: 'https://tracker.adobe.com/#/view/FP-4199039'.