1211099 - Valence just resets when turning on "Safari on iOS" runtime, never connects to plugged in iPad

Status: RESOLVED WONTFIX

(DevTools :: about:debugging, defect, P3)

Description

[donrhummy]

User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:41.0) Gecko/20100101 Firefox/41.0
Build ID: 20150929144111

Steps to reproduce:

Plugged iPad (trusted device/computer) into Linux and also (later) Windows computer. Opened WebIDE from Firefox 41. Selected from pulldown "Safari, Firefox and other webviews from iOS".


Actual results:

It showed a "loading" circle for half a second and then went back to the blank state.


Expected results:

It should have turned on and connected to my iPad (which also had Safari open). I have Google's ios_webkit_debug_proxy and that works, but the WebIDE tool does not.

Comment 1

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

What version of iOS is on the device?

What version of Valence (from about:addons) do you have?

Check the Browser Console[1] when connecting.  Do you see any errors?

[1]: https://developer.mozilla.org/en-US/docs/Tools/Browser_Console

Comment 2

[donrhummy]

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #1)
> What version of iOS is on the device?
> 
> What version of Valence (from about:addons) do you have?
> 
> Check the Browser Console[1] when connecting.  Do you see any errors?
> 
> [1]: https://developer.mozilla.org/en-US/docs/Tools/Browser_Console

iOS 8.4
Valence 0.3.2

How do I check the console? WebIDE is a popup window and pressing ctrl+shift+j does nothing.

Comment 3

[donrhummy]

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #1)
> What version of iOS is on the device?
> 
> What version of Valence (from about:addons) do you have?
> 
> Check the Browser Console[1] when connecting.  Do you see any errors?
> 
> [1]: https://developer.mozilla.org/en-US/docs/Tools/Browser_Console

OK, I opened the console via another web page and then went back to WebIDE and I get:

Error: 'unwatchApps' request packet has no destination.: DebuggerClient.prototype.request@resource://gre/modules/devtools/dbg-client.jsm:737:1
AppActorFront.prototype._unlistenAppEvents@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/app-actor-front.js:672:12
AppActorFront.prototype.unwatchApps@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/app-actor-front.js:735:12
Monitor.disconnectFromRuntime@chrome://webide/content/monitor.js:149:7
Monitor.onAppManagerUpdate@chrome://webide/content/monitor.js:124:11
EventEmitter_emit@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/event-emitter.js:147:11
exports.AppManager.update@resource://gre/modules/commonjs/toolkit/loader.js -> resource:///modules/devtools/webide/app-manager.js:143:5
exports.AppManager.onConnectionChanged@resource://gre/modules/commonjs/toolkit/loader.js -> resource:///modules/devtools/webide/app-manager.js:204:5
EventEmitter_emit@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/event-emitter.js:147:11
Connection.prototype._setStatus@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/client/connection-manager.js:344:5
Connection.prototype._onDisconnected@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/client/connection-manager.js:368:5
eventSource/aProto.addOneTimeListener/l@resource://gre/modules/devtools/dbg-client.jsm:137:7
eventSource/aProto.emit@resource://gre/modules/devtools/dbg-client.jsm:194:9
DebuggerClient.prototype.onClosed@resource://gre/modules/devtools/dbg-client.jsm:1112:5
LocalDebuggerTransport.prototype.close@resource://gre/modules/devtools/dbg-client.jsm -> resource://gre/modules/devtools/transport/transport.js:666:9
LocalDebuggerTransport.prototype.close@resource://gre/modules/devtools/dbg-client.jsm -> resource://gre/modules/devtools/transport/transport.js:662:7
LocalDebuggerTransport.prototype.close@resource://gre/modules/devtools/dbg-client.jsm -> resource://gre/modules/devtools/transport/transport.js:662:7
DebuggerClient.prototype.close/detachClients@resource://gre/modules/devtools/dbg-client.jsm:442:9
DebuggerClient.prototype.close@resource://gre/modules/devtools/dbg-client.jsm:456:5
Connection.prototype.disconnect@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/client/connection-manager.js:260:9
TabStore.prototype.listTabs/<@resource://gre/modules/commonjs/toolkit/loader.js -> resource:///modules/devtools/webide/tab-store.js:104:9
DebuggerClient.requester/</<@resource://gre/modules/devtools/dbg-client.jsm:352:9
makeInfallible/<@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/DevToolsUtils.js:83:14
emitOnObject@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/commonjs/sdk/event/core.js:112:9
emit@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/commonjs/sdk/event/core.js:89:38
Request.prototype.emit@resource://gre/modules/devtools/dbg-client.jsm:1225:29
DebuggerClient.prototype.onPacket/emitReply@resource://gre/modules/devtools/dbg-client.jsm:1036:29
DebuggerClient.prototype.onPacket@resource://gre/modules/devtools/dbg-client.jsm:1038:9
LocalDebuggerTransport.prototype.send/<@resource://gre/modules/devtools/dbg-client.jsm -> resource://gre/modules/devtools/transport/transport.js:569:11
makeInfallible/<@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/DevToolsUtils.js:83:14
makeInfallible/<@resource://gre/modules/commonjs/toolkit/loader.js -> resource://gre/modules/devtools/DevToolsUtils.js:83:14
 event-emitter.js:152:0
Connection status changed: connecting app-manager.js:162
Connection status changed: connected app-manager.js:162

The last "connected" message - I don't believe. It never does anything.

Comment 4

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

donrhummy, are you running ios_webkit_debug_proxy yourself somewhere in the background?  You should not need to do this, Valence will you start it for you, and the other one might conflict.

Perhaps something strange is happening when WebIDE tries to "auto-connect" to the last device?  That's what the error above makes me think of...  You can go to Menu Bar -> Project -> Preferences and disable "Reconnect to previous runtime" in WebIDE.

I no longer have an iOS 8.4 device myself, so I can only really test from simulators.  :past, do you still have a iOS 8.4 device to test with?

Comment 5

[donrhummy]

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #4)
> donrhummy, are you running ios_webkit_debug_proxy yourself somewhere in the
> background?  

No. I run ios_webkit_debug_proxy on a different computer and that one works perfectly. This computer only has WebIDE/Valence.

> Perhaps something strange is happening when WebIDE tries to "auto-connect"
> to the last device?  That's what the error above makes me think of...  You
> can go to Menu Bar -> Project -> Preferences and disable "Reconnect to
> previous runtime" in WebIDE.
> 
> I no longer have an iOS 8.4 device myself, so I can only really test from
> simulators.  :past, do you still have a iOS 8.4 device to test with?

yes I do.

Comment 6

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

(In reply to donrhummy from comment #5)
> > I no longer have an iOS 8.4 device myself, so I can only really test from
> > simulators.  :past, do you still have a iOS 8.4 device to test with?
> 
> yes I do.

This bit was meant for the other person I've flagged, so I'm hoping he may be able to help here.

Comment 7

[donrhummy]

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #6)
> (In reply to donrhummy from comment #5)
> > > I no longer have an iOS 8.4 device myself, so I can only really test from
> > > simulators.  :past, do you still have a iOS 8.4 device to test with?
> > 
> > yes I do.
> 
> This bit was meant for the other person I've flagged, so I'm hoping he may
> be able to help here.

np. I also tried your other suggestion in preferences and it didn't fix the problem. Still occurs.

Comment 8

[Panos Astithas (he/him) [:past] (please ni?)]

I've tested on iOS 8.4.1 and I got a successful debugging session on an older version of Developer Edition. In latest Nightly and Developer Edition I get these errors, but the connection appears to be established (no tabs to debug though):

1444493290295	addons.xpi	WARN	Add-on [e-mail] is missing bootstrap method startup
* daemon not running. starting it now on port 5037 *
* daemon started successfully *
1444493301977	addons.xpi	WARN	Add-on [e-mail] is missing bootstrap method shutdown

Comment 9

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

:past, thanks for testing!  Just to clarify, you're saying with Nightly, you do connect, but see no tabs, whereas with old Dev. Ed, you connect *and* see tabs (as expected)?

If that's correct, it might be related to the WebIDE change to sidebars, which has altered when we request the tab list somewhat.

Comment 10

[Panos Astithas (he/him) [:past] (please ni?)]

That's correct (and unfortunately I no longer have access to said iOS device).

Comment 11

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

I think this issue will be improved by bug 1210123, which will make the tabs refresh button always on screen to ensure you have the latest tab list.

That said, I don't have a physical iOS 8.4 device (just simulators), so it is hard for me to be sure.

Related to this topic, there appear to be further complications on iOS 9, which are tracked over on GitHub[1].

[1]: https://github.com/mozilla/valence/issues/199

Comment 12

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

I just now had some trouble with 8.4.1 simulator not showing the tabs, but was able to resolve it by refreshing the tabs (by hacking the DOM to show me the button).

So, I am now more confident that bug 1210123 will help.

Comment 13

[donrhummy]

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #12)
> I just now had some trouble with 8.4.1 simulator not showing the tabs, but
> was able to resolve it by refreshing the tabs (by hacking the DOM to show me
> the button).
> 
> So, I am now more confident that bug 1210123 will help.

You *might* be correct, but why am I getting different error messages (comment #3) than you?

Comment 14

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

(In reply to donrhummy from comment #13)
> (In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #12)
> > I just now had some trouble with 8.4.1 simulator not showing the tabs, but
> > was able to resolve it by refreshing the tabs (by hacking the DOM to show me
> > the button).
> > 
> > So, I am now more confident that bug 1210123 will help.
> 
> You *might* be correct, but why am I getting different error messages
> (comment #3) than you?

I agree, they may be different errors.  I am so far unsure about your case, as I haven't seen that precise stack trace locally yet.  It may take some more investigation to find out each case.  If they are separate, we can break them out into separate bugs or something.

If you try using Nightly 44, do you see any difference in behavior?

Comment 15

[donrhummy]

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #14)
> (In reply to donrhummy from comment #13)
> > (In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #12)
> > > I just now had some trouble with 8.4.1 simulator not showing the tabs, but
> > > was able to resolve it by refreshing the tabs (by hacking the DOM to show me
> > > the button).
> > > 
> > > So, I am now more confident that bug 1210123 will help.
> > 
> > You *might* be correct, but why am I getting different error messages
> > (comment #3) than you?
> 
> I agree, they may be different errors.  I am so far unsure about your case,
> as I haven't seen that precise stack trace locally yet.  It may take some
> more investigation to find out each case.  If they are separate, we can
> break them out into separate bugs or something.
> 
> If you try using Nightly 44, do you see any difference in behavior?

How do I run FF nightly on a linux desktop alongside the current FF? When I downloaded and extracted nightly and clicked the "firefox" executable, it just opened a new window in the current FF.

Comment 16

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

(In reply to donrhummy from comment #15)
> How do I run FF nightly on a linux desktop alongside the current FF? When I
> downloaded and extracted nightly and clicked the "firefox" executable, it
> just opened a new window in the current FF.

Firefox uses a profile to store your user data, and only one Firefox version can access your profile at a time.  One option is to close all Firefox windows from the release version before starting Nightly, which should allow Nightly to startup with your default profile.  However, it's possible this could lead to some data loss when you later go back to release Firefox with that profile (moving a profile from a newer Firefox back to an older Firefox is not a supported path).  Many of us do it anyway, but there is some amount of risk.

Another option is create a separate profile for use with Nightly only.  You can do that by passing the -P option when starting Nightly from the terminal.  This will display a prompt to allow you to choose which profile to use and also allow you to create a new one for Nightly to use.

Comment 17

[donrhummy]

(In reply to J. Ryan Stinnett [:jryans] (use ni?) from comment #16)
> (In reply to donrhummy from comment #15)
> Another option is create a separate profile for use with Nightly only.  You
> can do that by passing the -P option when starting Nightly from the
> terminal.  This will display a prompt to allow you to choose which profile
> to use and also allow you to create a new one for Nightly to use.

Using Nightly, I now get this response/error:

    Connection status changed: connecting app-manager.js:161
    Connection status changed: connected app-manager.js:161
     protocol.js:907
     protocol.js:907
    Connection status changed: disconnecting app-manager.js:161
    Connection status changed: disconnected app-manager.js:161
    unknownError Promise-backend.js:937

Comment 18

[donrhummy]

When I clicked the "ios" button again, I got a new error:

valence:ios_webkit_debug_proxy: /home/myusername/Downloads/firefox/profiles/extensions/[e-mail]/tools/linux64/ios_webkit_debug_proxy: /lib64/libssl.so.1.0.0: no version information available (required by /home/myusername/Downloads/firefox/profiles/extensions/[e-mail]/tools/linux64/libimobiledevice.so.5)
/home/myusername/Downloads/firefox/profiles/extensions/[e-mail]/tools/linux64/ios_webkit_debug_proxy: /lib64/libcrypto.so.1.0.0: no version information available (required by /home/myusername/Downloads/firefox/profiles/extensions/[e-mail]/tools/linux64/libimobiledevice.so.5)
No device found, is it plugged in?

I then clicked it three more times, but don't get that again. It's very inconsistent

Comment 19

[Albert Scheiner [:alberts]]

not sure if I landed in the right bug, but trying to connect to an iPhone (4, iOS 7.1.2) or and iPad (2, iOS 9.2) with Firefox Dev 44.0a2 (2015-12-14); I can see the button for "WebViews on iOS" when one of them is connected to the PC (Win 7) > clicking the button shows the device connected, but that's that, the main window doesn't change and I see no option to select (a) tab(s) (anymore).

I see this error in the browser console: "valence:PowerShell -Command ps ios-webkit-debug-proxy: W" when the connection gets established

Comment 20

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

(In reply to Albert Scheiner [:alberts] from comment #19)
> not sure if I landed in the right bug, but trying to connect to an iPhone
> (4, iOS 7.1.2) or and iPad (2, iOS 9.2) with Firefox Dev 44.0a2
> (2015-12-14); I can see the button for "WebViews on iOS" when one of them is
> connected to the PC (Win 7) > clicking the button shows the device
> connected, but that's that, the main window doesn't change and I see no
> option to select (a) tab(s) (anymore).
> 
> I see this error in the browser console: "valence:PowerShell -Command ps
> ios-webkit-debug-proxy: W" when the connection gets established

It could be this bug, which I am so far unsure of the root cause.

Additionally there is https://github.com/mozilla/valence/issues/199 which is specific to iOS 9 devices, unlike this bug which also has trouble with earlier devices.

Comment 21

[Julian Descottes [:jdescottes]]

About:debugging bug triage. Filter on TRIAGE-JD201705

Temporarily setting P3 to bugs migrated from webide to cleanup the triage for about:debugging.

Comment 22

[J. Ryan Stinnett [:jryans] (Use needinfo, replies may be slow)]

Valence is no longer supported.