Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Electron broken on OS X in Apple Sandboxed apps (App Store) #3871

Closed
sethlu opened this issue Dec 19, 2015 · 109 comments · Fixed by #5205
Closed

Electron broken on OS X in Apple Sandboxed apps (App Store) #3871

sethlu opened this issue Dec 19, 2015 · 109 comments · Fixed by #5205

Comments

@sethlu
Copy link
Contributor

sethlu commented Dec 19, 2015

🍎 Not sure if this is related to the Chromium itself.
I've been trying and successfully packed and signed my app, but still there are logs in console like:

12/19/15 9:45:14.000 PM kernel[0]: Sandbox: Electron Helper(15181) deny(1) mach-lookup org.chromium.Chromium.rohitfork.15175
12/19/15 9:45:14.000 PM kernel[0]: Sandbox: Electron Helper(15181) deny(1) mach-lookup org.chromium.Chromium.iosurfacemgr.15175
12/19/15 9:45:15.520 PM sandboxd[326]: ([15176]) Electron Helper(15176) deny mach-lookup org.chromium.Chromium.rohitfork.15175
12/19/15 9:45:15.574 PM sandboxd[326]: ([15176]) Electron Helper(15176) deny mach-lookup org.chromium.Chromium.iosurfacemgr.15175
12/19/15 9:45:15.926 PM sandboxd[326]: ([15177]) Electron Helper(15177) deny mach-lookup org.chromium.Chromium.rohitfork.15175
12/19/15 9:45:15.958 PM sandboxd[326]: ([15177]) Electron Helper(15177) deny mach-lookup org.chromium.Chromium.iosurfacemgr.15175
12/19/15 9:45:16.742 PM sandboxd[326]: ([15178]) Electron Helper(15178) deny mach-lookup org.chromium.Chromium.rohitfork.15175
12/19/15 9:45:16.773 PM sandboxd[326]: ([15178]) Electron Helper(15178) deny mach-lookup org.chromium.Chromium.iosurfacemgr.15175
12/19/15 9:45:16.946 PM sandboxd[326]: ([15180]) Electron Helper(15180) deny mach-lookup org.chromium.Chromium.rohitfork.15175
12/19/15 9:45:16.977 PM sandboxd[326]: ([15180]) Electron Helper(15180) deny mach-lookup org.chromium.Chromium.iosurfacemgr.15175

The following extract from my entitlements doesn't seem work work. I guess that Wildcards may not be accepted; however, the * bits keep changing every time I execute the application.

<key>com.apple.security.temporary-exception.mach-lookup.global-name</key>
<array>
  <string>org.chromium.Chromium.rohitfork.*</string>
  <string>org.chromium.Chromium.iosurfacemgr.*</string>
</array>

May there be any fixes on this in future releases?
P.S.: I've found that the sandboxed app runs much slower and has some issues (like jittering graphics, the best way I could describe it). Again, I guess it is related to the denials from sandboxing.
Thanks!

@sethlu
Copy link
Contributor Author

sethlu commented Dec 19, 2015

The Electron version was 0.36.1 MAS build.
I've tested with the 0.36.0 OS X version; sadly the same occurred.
Thanks again.

@sethlu
Copy link
Contributor Author

sethlu commented Feb 12, 2016

On this issue, I've already tested with v0.35.x (mas builds). Though still giving the deny mach-lookup logs, the graphics issue doesn't seem to occur at all.
Also need to mention: For the mas builds (tested with v0.36.0-v.36.7 mas builds), before signing with sandbox entitlements the application works fine. The graphics issue only appears after code-signed.
The darwin builds are not affected, as they do not get signed with entitlements for sandboxing.

@alchen
Copy link

alchen commented Feb 12, 2016

Same here. I'm reverting back to v0.35.x for now. Maybe we need to wait until the upgrade to Chrome 48 over at electron/libchromiumcontent#175 arrives.

@matthieualouis
Copy link

I have the same issue, and I'm using the very latest MAS build (0.37.1). This latest build upgrades to Chromium 49 so I thought it might resolve this, but it doesn't.

@sethlu
Copy link
Contributor Author

sethlu commented Mar 14, 2016

Just tested some of the recent releases of Electron:

  • Electron v0.36.x darwin builds all work very well.
  • Electron v0.36.x mas builds work with some glitch in graphics after sandboxed (I assume with CSS transitions and animations, or for things related to GPU)?
  • Electron v0.37.0 darwin/mas fails to start (fixed in v0.37.1).
  • Electron v0.37.1 darwin build has some glitch like that in Electron v0.36.x mas builds (before and after code signed for distribution).
  • Electron v0.37.1 mas build has some glitch like that in Electron v0.36.x mas builds; however, it fails to load after sandboxed?

@sethlu
Copy link
Contributor Author

sethlu commented Mar 14, 2016

(cont'd) when signing with only sandbox entitlements for Electron v0.37.1 mas build:

3/14/16 5:23:51.171 PM sandboxd[395]: ([3742]) test-0.37.1 Help(3742) deny mach-lookup org.chromium.Chromium.rohitfork.3741
3/14/16 5:23:51.311 PM sandboxd[395]: ([3743]) test-0.37.1 Help(3743) deny mach-lookup org.chromium.Chromium.rohitfork.3741
3/14/16 5:23:51.461 PM sandboxd[395]: ([3744]) test-0.37.1 Help(3744) deny mach-lookup org.chromium.Chromium.rohitfork.3741
3/14/16 5:23:51.719 PM Electron[3741]: __net_helper_get_connection_block_invoke_3 could not connect to networkd
3/14/16 5:23:51.719 PM Electron[3741]: __nw_path_evaluator_start_helper_connection_block_invoke net_helper_path_evaluation_start callback failed, dumping backtrace:
        [x86_64] libnetcore-583.20.10
    0   libsystem_network.dylib             0x00007fff96104ba5 __nw_create_backtrace_string + 123
    1   libsystem_network.dylib             0x00007fff9611de68 __nw_path_evaluator_start_helper_connection_block_invoke + 22
    2   libxpc.dylib                        0x00007fff8c19691f _xpc_connection_reply_callout + 26
    3   libxpc.dylib                        0x00007fff8c1968c0 _xpc_connection_call_reply + 36
    4   libdispatch.dylib                   0x00007fff9b90b33f _dispatch_client_callout + 8
    5   libdispatch.dylib                   0x00007fff9b90ff6f _dispatch_queue_drain + 754
    6   libdispatch.dylib                   0x00007fff9b91663b _dispatch_queue_invoke + 549
    7   libdispatch.dylib                   0x00007fff9b90ec87 _dispatch_root_queue_drain + 538
    8   libdispatch.dylib                   0x00007fff9b90ea34 _dispatch_worker_thread3 + 91
    9   libsystem_pthread.dylib             0x00007fff91b6768f _pthread_wqthread + 1129
    10  libsystem_pthread.dylib             0x00007fff91b65365 start_wqthread + 13
3/14/16 5:23:51.720 PM Electron[3741]: nw_path_evaluator_start_helper_connection net_helper_path_evaluation_start failed, dumping backtrace:
        [x86_64] libnetcore-583.20.10
    0   libsystem_network.dylib             0x00007fff96104ba5 __nw_create_backtrace_string + 123
    1   libsystem_network.dylib             0x00007fff9611a228 nw_path_evaluator_start_helper_connection + 196
    2   libdispatch.dylib                   0x00007fff9b916871 _dispatch_call_block_and_release + 12
    3   libdispatch.dylib                   0x00007fff9b90b33f _dispatch_client_callout + 8
    4   libdispatch.dylib                   0x00007fff9b90ff6f _dispatch_queue_drain + 754
    5   libdispatch.dylib                   0x00007fff9b91663b _dispatch_queue_invoke + 549
    6   libdispatch.dylib                   0x00007fff9b90ec87 _dispatch_root_queue_drain + 538
    7   libdispatch.dylib                   0x00007fff9b90ea34 _dispatch_worker_thread3 + 91
    8   libsystem_pthread.dylib             0x00007fff91b6768f _pthread_wqthread + 1129
    9   libsystem_pthread.dylib             0x00007fff91b65365 start_wqthread + 13
3/14/16 5:23:51.917 PM sandboxd[395]: ([3741]) Electron(3741) deny mach-lookup com.apple.network

I thought Electron may try to load up remote resources, so then added entitlements (still with sandbox) for network client:

3/14/16 5:27:58.223 PM sandboxd[395]: ([3795]) test-0.37.1 Help(3795) deny mach-lookup org.chromium.Chromium.rohitfork.3794
3/14/16 5:27:58.378 PM sandboxd[395]: ([3796]) test-0.37.1 Help(3796) deny mach-lookup org.chromium.Chromium.rohitfork.3794
3/14/16 5:27:58.523 PM sandboxd[395]: ([3797]) test-0.37.1 Help(3797) deny mach-lookup org.chromium.Chromium.rohitfork.3794

Interesting, the logs about networking disappeared. However, the app still doesn't load properly with something blank as attached.

screenshot 2016-03-14 17 29 01

Looks a bit different from #3771.

@hachi8833
Copy link

Same issue on 0.37.1.

Looks like the issue comes from the recent CEF builds:
http://www.magpcss.org/ceforum/viewtopic.php?f=6&t=13469

FYI: prepared the two Electron-MAS binaries to reproduce the issue.
https://drive.google.com/folderview?id=0B8jpVsBA-4FrOTllTlZGOU5JeFU&usp=sharing

Electron_v0.35.6-mas.app can be launched while Electron_v0.37.2-mas.app does not work. The only difference between them is MAS version.

FYI2: FYI https://github.com/electron-userland/electron-osx-sign/wiki/2.-Electron-Compatibility

@sethlu
Copy link
Contributor Author

sethlu commented Mar 16, 2016

Just tested; the same from Electron v0.37.1 appears in v0.37.2.

@herrmannplatz
Copy link
Contributor

Can confirm this for 0.37.2

@sethlu
Copy link
Contributor Author

sethlu commented Mar 18, 2016

Seems that the launching issue in v0.37.1's been fixed in v0.37.2. However, v0.37.2 mas seems failing when sandboxed; it freezes with the window showing up.

@jarek-foksa
Copy link

Using v0.37.2 with MAS sandbox, I can create new browser windows, but the app freezes after I call browserWindow.loadURL(someURL). There are no errors printed in the console.

@fuermosi777
Copy link

v0.36.2 works perfect after signing, but after signing v0.37.2, the app keeps showing "Not responding" and no window showed up.

@luin
Copy link

luin commented Mar 25, 2016

I got the same issue as @fuermosi777. The signed app (v0.37.2) kept not responding while it worked perfect with v0.35.4.

Related logs here: https://gist.github.com/luin/33074d4f340aba0cc000

@tosslab-jay
Copy link

Hi, I'm a product manager at JANDI, a business messaging platform from South Korea, which is being used by dozens of thousands active users everyday. We're trying to shift to Electron for our web app but this particular issue is a major road block for us for releasing to MAS. I really wish this issue can be solved in a near future so that we can add my product to the list of Electron-used services. Always appreciating the hard work of the crew. Thanks.

@sethlu
Copy link
Contributor Author

sethlu commented Mar 29, 2016

Just tested; the same from Electron v0.37.1 appears in v0.37.2.

Hey, just some updates with my testing with v0.37.3:

Before sandboxing/codesigning:

  • darwin build works, with some slight graphics issues.
  • mas build the same as above.

After sandboxing/codesigning:

  • darwin build has some graphics issue drawing the elements when animating, as far as I've tested.
  • mas build doesn't really load up the webpage, and freezes. It doesn't crash by itself, needs terminating.

I'm thinking about changing a title for this issue, as rohitfork and iosurfacemgr aren't the real factors causing these issues. 😕

@tyv
Copy link

tyv commented Apr 2, 2016

same problem, looking forward for fix

@tyv
Copy link

tyv commented Apr 2, 2016

sad thing I need API introduced at 0.36.11 and can't just use old mas build.
any workaround ideas?

@anaisbetts anaisbetts changed the title rohitfork and iosurfacemgr issues after app sandboxed Electron broken on OS X in Apple Sandboxed apps (App Store) Apr 2, 2016
@anaisbetts
Copy link
Contributor

This is a super important bug for us at Slack, we'll probably have a look at this soon

@zcbenz
Copy link
Member

zcbenz commented Apr 7, 2016

For people interested in fixing this, here are some hints:

  • The root cause is after sandboxing Chromium's IPC system starts to have troubles, in 0.36 GPU process is unable to start, and in 0.37 even renderer process is not able to run.
  • This starts to happen from 0.36, which upgrades Chromium from 45 to 47.
  • I believe this is because Chromium started to use XPC instead of raw mach messages for IPC.

Related issues in Chromium:
https://bugs.chromium.org/p/chromium/issues/detail?id=382931
https://bugs.chromium.org/p/chromium/issues/detail?id=367863

@zcbenz
Copy link
Member

zcbenz commented Apr 7, 2016

A possible solution is to force using old system by setting is_yosemite_or_later_ to false in PreExecDelegate::PreExecDelegate:
https://code.google.com/p/chromium/codesearch#chromium/src/sandbox/mac/pre_exec_delegate.cc&sq=package:chromium&type=cs&l=37

But from the Chromium issue they switched to use XPC because from Yosemite the old implemented is not compatible with the OS and crash happens.

@tyv
Copy link

tyv commented Jul 25, 2016

@tyv
Copy link

tyv commented Jul 25, 2016

the main concern is, after I sign app (in different ways without any errors), it is unable to run it.
it self closes itself.

@sethlu
Copy link
Contributor Author

sethlu commented Jul 25, 2016

@tyv, I think your app is well signed. May you check in the Console as well to see if any error arises when your app launches? Also, I would recommend removing the temporary exception in https://github.com/ubergrape/grape-electron/blob/master/resources/osx/child.plist#L9-L10 before building another testing release.

@tyv
Copy link

tyv commented Jul 26, 2016

@sethlu
i see only

7/26/16 13:36:49.427 sandboxd[19719]: ([20302]) Grape(20302) deny network-bind /private/var/folders/dr/cw1hncvj5vz4_p44q4nbjzgm0000gn/T/com.ChatGrape/S/SS

@tyv
Copy link

tyv commented Jul 26, 2016

@tyv
Copy link

tyv commented Jul 26, 2016

and one more

7/26/16 13:54:54.241 taskgated[509]: no application identifier provided, can't use provisioning profiles [pid=20987]

@tyv
Copy link

tyv commented Jul 29, 2016

@sethlu it looks like it is same as here: electron/osx-sign/issues/59
any update on that problem?

@sethlu
Copy link
Contributor Author

sethlu commented Aug 1, 2016

Hi @tyv, apologies for my late reply; I've been away for the past few days.

The issue electron/osx-sign#59 I think is slightly different in the way that a critical problem lies on the video en/decoder. (However, I don't know how @jpittner got around the video en/decoder issue... Or just following the conceptual signing flow at https://github.com/sethlu/electron-distribution-guide then everything just naturally resolved.)

The app should still launch with no application identifier provided... and without significant effects as far as I have heard 😕. So far I believe with your app correctly signed, the only signage is really only deny network-bind... Let me have a check for a resolution.

@jpittner
Copy link
Contributor

jpittner commented Aug 1, 2016

@the video decoder/encoder stuff went away once the package was properly signed. Right now, I've got it working with MAS build (and can use a development provisioning profile to test it) but am having some trouble getting the developer-id signed version (that is, for non-MAS distribution) to work properly!

edit - spoke too soon. That was with prior version (1.2.x) with 1.3.1 I'm now running into problems getting the MAS build to load up on my development machine.

@sethlu
Copy link
Contributor Author

sethlu commented Aug 5, 2016

@tyv I've managed to investigate a bit more over the past few days... Apple sandbox does not allow network-bind outside sandboxed containers but I am not sure what actually caused your app to have such behavior. Have you tried to create sockets in your code?


@jpittner Thanks for the info! I will have a look at the latest MAS builds.

@kof
Copy link

kof commented Aug 5, 2016

@sethlu by sockets you also include websocket?

@tyv
Copy link

tyv commented Aug 5, 2016

@sethlu

Have you tried to create sockets in your code?

all this app does is change location.href to the https://... where is web app is and this app yes, opens websocket connection.

@sethlu
Copy link
Contributor Author

sethlu commented Aug 5, 2016

@kof About web socket, from the text network-bind I think it does not matter much as long as it does not touch critical areas about binding essentially I guess?

Network-bind
Syntax: Actions: Filters: Modifiers:
(action network-bind [filter] [modifier]) allow deny
network path file-mode
send-signal no-log
Definition:
Control access to socket bind(). It has no support for IP filtering, it must be localhost or * (check network filter for more information).
Applies to:
bind, unp_bind
Example(s):
(deny network-bind (local ip "*:7890"))
$ sandbox-exec -f ls2 nc -l 7890
nc: Operation not permitted
Log output:
Sep 2 21:08:41 macbox sandboxd[45438]: nc(45437) deny network-bind 0.0.0.0:7890

Ref: https://reverse.put.as/wp-content/uploads/2011/09/Apple-Sandbox-Guide-v1.0.pdf

@kof
Copy link

kof commented Aug 5, 2016

so maybe it is the web sockets from non-localhost domain?

@sethlu
Copy link
Contributor Author

sethlu commented Aug 5, 2016

@tyv Have you tried to add com.apple.security.network.client in the app(/parent) entitlements file? Sorry I'm not sure if it may work for that I have not encountered similar cases.

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
  <dict>
    <key>com.apple.security.app-sandbox</key>
    <true/>
    <key>com.apple.security.network.client</key>
    <true/>
    <key>com.apple.security.application-groups</key>
    <string>Y8DPE6DGC7.com.ChatGrape</string>
  </dict>
</plist>

@tyv
Copy link

tyv commented Aug 5, 2016

I'll try

@tyv
Copy link

tyv commented Aug 5, 2016

I am able to run empty app after sign.
Will try to find what cause problem and will report back.

@tyv
Copy link

tyv commented Aug 5, 2016

okay, I found what kills app when it is signed
http://electron.atom.io/docs/all/#appmakesingleinstancecallback
shouldQuit === true in this example from old official doc:

const shouldQuit = app.makeSingleInstance(() => {
  const {mainWindow} = state
  // Someone tried to run a second instance, we should focus our window
  if (mainWindow) showMainWindow()
  return true
})

if (shouldQuit) quit()

@tyv
Copy link

tyv commented Aug 5, 2016

@sethlu it is same with new snippet from the doc

@sethlu
Copy link
Contributor Author

sethlu commented Aug 7, 2016

@tyv does you app work now?

@tyv
Copy link

tyv commented Aug 7, 2016

I removed that code and yes, but yet didn't have time to check whats wrong with that method
why it is always returns true when app is signed

On Aug 7, 2016, at 12:06, Zhuo Lu notifications@github.com wrote:

@tyv does you app work now?


You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or mute the thread.

@sethlu
Copy link
Contributor Author

sethlu commented Aug 7, 2016

@tyv A couple of places I think where this issue occurs:

  1. switch (process_singleton_->NotifyOtherProcessOrCreate()) {
    case ProcessSingleton::NotifyResult::LOCK_ERROR:
    case ProcessSingleton::NotifyResult::PROFILE_IN_USE:
    case ProcessSingleton::NotifyResult::PROCESS_NOTIFIED:
    process_singleton_.reset();
    return true;
  2. ProcessSingleton::NotifyOtherProcessWithTimeoutOrCreate(
    const base::CommandLine& command_line,
    int retry_attempts,
    const base::TimeDelta& timeout) {
    NotifyResult result = NotifyOtherProcessWithTimeout(
    command_line, retry_attempts, timeout, true);
    if (result != PROCESS_NONE)
    return result;
    if (Create())
    return PROCESS_NONE;
    // If the Create() failed, try again to notify. (It could be that another
    // instance was starting at the same time and managed to grab the lock before
    // we did.)
    // This time, we don't want to kill anything if we aren't successful, since we
    // aren't going to try to take over the lock ourselves.
    result = NotifyOtherProcessWithTimeout(
    command_line, retry_attempts, timeout, false);
    if (result != PROCESS_NONE)
    return result;
    return LOCK_ERROR;
    }
  3. ProcessSingleton::NotifyResult ProcessSingleton::NotifyOtherProcessWithTimeout(
    const base::CommandLine& cmd_line,
    int retry_attempts,
    const base::TimeDelta& timeout,
    bool kill_unresponsive) {
    DCHECK_GE(retry_attempts, 0);
    DCHECK_GE(timeout.InMicroseconds(), 0);
    base::TimeDelta sleep_interval = timeout / retry_attempts;
    ScopedSocket socket;
    for (int retries = 0; retries <= retry_attempts; ++retries) {
    // Try to connect to the socket.
    if (ConnectSocket(&socket, socket_path_, cookie_path_))
    break;

Seems that Electron does open a socket and send out a message checking process singleton... I'm not yet sure how to avoid having the network binding be addressed in sandbox though.

@tyv
Copy link

tyv commented Aug 7, 2016

okay, i see, so if this isn't fixable, probably comment in guide is a good idea.

@tyv
Copy link

tyv commented Aug 7, 2016

or default value should be false, and still comment near the method doc, would be nice

@anaisbetts
Copy link
Contributor

Why would this not be fixable? It's our code - can you file a separate bug?

@tyv
Copy link

tyv commented Aug 8, 2016

@paulcbetts here you go ^^^

@patricksebastien
Copy link

Hi,
Not fix at all, even with a simple HelloWorld. Anyone have an idea of what needs to be done to sumbit my app to Mac Store (the build is working nicely when not signing the app).

electron-packager . "MyApp" --platform=mas --arch=x64 --overwrite --version=1.4.4 --app-bundle-id="com.mycie.myapp" --app-version="1.0.0" --build-version="666"

electron-osx-sign "MyApp.app" --entitlements="mas.default.entitlements" --entitlements-inherit="mas.inherit.default.entitlements"

electron-osx-flat "MyApp"

mas.default.entitlements

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
  <dict>
    <key>com.apple.security.app-sandbox</key>
    <true/>
    <key>com.apple.security.network.client</key>
    <true/>
    <key>com.apple.security.network.server</key>
    <true/>
    <key>com.apple.security.application-groups</key>
    <string>XXXXXXXXXXX.com.mycie.myapp</string>
  </dict>
</plist>

mas.inherit.default.entitlements

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
  <dict>
    <key>com.apple.security.app-sandbox</key>
    <true/>
    <key>com.apple.security.inherit</key>
    <true/>
  </dict>
</plist>
electron-osx-sign:warn No `platform` passed in arguments, checking Electron platform... +0ms
  electron-osx-sign:warn No `identity` passed in arguments, discovering identities... +3ms
  electron-osx-sign Signing application... +1s
  electron-osx-sign > application         MyApp-mas-x64/MyApp.app +0ms
  electron-osx-sign > platform            mas +0ms
  electron-osx-sign > entitlements        ../mas.default.entitlements +0ms
  electron-osx-sign > child-entitlements  ../mas.inherit.default.entitlements +0ms
  electron-osx-sign > additional-binaries  +1ms
  electron-osx-sign > identity            3rd Party Mac Developer Application: BAYARD PRESSE (VE4JH2R5J4) +0ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/MyApp Helper EH.app/Contents/MacOS/MyApp Helper EH +117ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/MyApp Helper EH.app +130ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/MyApp Helper NP.app/Contents/MacOS/MyApp Helper NP +120ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/MyApp Helper NP.app +119ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/MyApp Helper.app/Contents/MacOS/MyApp Helper +121ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/MyApp Helper.app +119ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/Electron Framework.framework/Versions/A/Electron Framework +122ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/Electron Framework.framework/Versions/A/Libraries/libffmpeg.dylib +1s
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/Electron Framework.framework/Versions/A/Libraries/libnode.dylib +152ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Frameworks/Electron Framework.framework +318ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/MacOS/MyApp +1s
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app/Contents/Resources/app/PepperFlashPlayer.plugin/Contents/MacOS/PepperFlashPlayer +599ms
  electron-osx-sign Signing... MyApp-mas-x64/MyApp.app +410ms
  electron-osx-sign Verifying sign... +612ms
  electron-osx-sign Verifying entitlements... +278ms
Application signed: MyApp-mas-x64/MyApp.app

@urgn-zz
Copy link

urgn-zz commented Jun 18, 2019

I'm facing this issue, what is the solution?
When
com.apple.security.app-sandbox is in my plist - my app gets whitescreen.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging a pull request may close this issue.