I develop a web browser based on gtk+ and webkit2gtk in Rust and sometimes, all GTK+ windows become black.
Even the gtk inspector window that we get with the environment variable GTK_DEBUG=interactive is black.
Even thought the windows are black, the UI is still responsive since I can navigate the web with the keyboard (I see the window title updating, showing the new page URL and load progress).
Here are two actions that triggers this issue everytime they happen:
Destroying the web view
When running the application a second time, it sends a message via Unix Domain Socket to the first process so that it creates a new web view.
I have unfortunately no small example to reproduce the issue. If you want, I can show you the code of the project, but it is big, non-trivial and uses many abstraction layers over gtk+.
I know I don't give you a lot to help me, but if you can give me some explanations about how the rendering works and how to debug that, it would be very appreciated.
Can you give me some hints on how to debug this issue?
Is there a global OpenGL (or whatever) context for the GTK+ windows?
Are there some debuging tools to help me? (Setting G_MESSAGES_DEBUG=all does not show anything relevant.)
With strace, I was able to debug this issue:
I found out the FD used for IPC communication was still polled after being closed, so it was returning POLLNVAL.
Removing the FD with g_source_remove_unix_fd() fixed this issue.
Related
I have encountered a critical issue that Unity Editor freezes. I've spent many hours debugging step by step of my codes but still can't find where the problem is. So I think maybe I should try thinking from another angle, generally speaking what reasons would cause Unity Editor freezes and not responding?
I can't find a general case discussion about this topic.
From my experience, infinite loop is one reason for sure. Deadlock is critical issue, but not sure if it causes Unity Editor freezes. Unity Editor bug that I encountered only makes the whole editor crashes, instead of freezing. Any other experiences are welcomed. Thank you!
In such case, what kind of tools or methods could I use to debug it? Right now since the editor freezes I can't use "print" to find out what happens after it freezes. So I use Visual studio to debug the Editor thread, in this way I can see all the prints that I wrote. It appears the game is still running, only the editor not responding. And I can use VS click "attach to Unity and play" and put some debugging points, then debug step by step.
The first thing that I would check out is for an infinite loop. At the hang/freeze moment, you can attatch the debugger of your choice and pause the execution. In the case that it is an infinite loop that it is executing, at the exec time pause you might find the execution in a forever running while (true) {...}
Other thing that I would check is the plugins in use. Several Unity plugins like Parse, FMOD, UMP (Universal Media Player), ZFBrowser, or Embedded Browser are using native threads. It’s an issue when a plugin ends up attaching a native thread to the runtime, which then does blocking calls to the OS. This means Unity can't interrupt that thread for the debugger (or domain reload) and hang. Source
To check that you can check the active threads in the visual studio command window af the freeze is reproduced:
View->OtherWindows->CommandWindow and type in this command:
Debug.ListCallStack /AllThreads /ShowExternalCode
In the stack you can check if some thread is there with no need, or if its related with the plugins mentioned above.
Also an interesting point is to check in the windows task manager (in the case that you are using windows) if the CPU usage is to 0%. It can lead you to the type of hang that is taking place.
Good luck.
Edit: I forgot to mention, you need to check also the unity logfiles
I notice this all the time, and its super frustrating.
Unfortunately, this could be any number of issues. I notice this issue most often when working in projects that are made for the Universal Windows Platform.
Try using the Task manager to monitor specific processes / threads running.
Some follow up questions:
What platform is your project currently targeting?
What version of Unity are you running? Have you tried other versions?
What are your computers specs? Is the OS up to date? Graphics Drivers?
Does it happen (or happen more often) when an external code editor is open? Perhaps try going to Preferences>External Editor > Regenerate Project files.
Are you using Unity Collab by chance? I've had issues where collab is stuck trying to communicate with Unity Servers / looking for changes. Try logging out of your Unity account through the editor, and log back in.
Have you tried looking for a Unity editor crash dump, or error log files? I think they can be found here C:\Users\username\AppData\LocalLow\Unity by default. Those files may give you more specific data concerning your problem.
Unity's new versions are getting more slower and slower. From my experience 2019 versions are the best and more stable.
I solved my issue. It's fundamentally an infinite loop.
It's not a simple case such as "while(true)". I'll try to explain.
My game was a PvP game, and I'm making a local AI. Usually my design pattern works fine, however I just turned off the simulation of "AI thinking time", and since the AI codes and server codes all run in local mode, the transmitting of data between server and client are replaced by local method call(meaning instantly executed before everything else).
There is a loophole in my server code. I use "Update" and a flag on server to change a specific game state, however in this particular case, it got into an infinite loop because the local method call is executed before the "Update". And because my AI now doesn't need real time to "think", it "acts" and transmits the event data to server right away. And since the transmitting doesn't need time any more, it calls the server method instantly, hence forming the infinite loop.
With Pry (but also with Rubymine), I'm trying to debug a certain point in the code (using binding.pry). After calling Capybara's save_screenshot, I'm unable to execute any Capybara related commands (all commands die on time-out). This works out of a "debug mode" and in other web-drivers like Poltergeist.
I took a couple of hours today trying to debug it. I think I found the problem - or at least a way around it.
Our web site has a couple of links that open content in another browser window. Since the automation is quite ancient, and in that time Selenium didn't have a decent way to switch window-context, what we do is to visit the opened page by URL, and by this keep only a single window open at any given time.
This works, but something strange happens when running this test on "debug mode" (using binding.pry for example). Right before we do any actions on that specific page, we take a screen-shot using Capybara's save_screenshot method. On debug this results a corrupted image, and any following Capybara methods will fail on time-out. Opening this page using the link, and handling the windows context switching with Capybara's handle_window method solves the issue. It's still a mystery why it only happens with Capybara wekit though (as other web-drivers work properly). I'm guessing that perhaps the DOM might be structured differently.
Bounty Update
I am leaving the question essentially the same, but I just want to point out at the beginning here that I'm most interested in help trying to get the socket between flash (Adobe Animate) and the native debugger working again, as I believe that is the issue I'm having. Or is it a socket between flash and Windows?
Original Question
I've apparently turned something off? Even when I write really bad code (like trying to call a function that isn't there or divide an integer into a fraction), the flash player boots, shows a background color and stops there. No messages in the output window or compiler errors window. If I fix my code, it all runs fine, but for about 30 minutes (ever since I started trying to work with bitmaps for a sprite sheet) I get no runtime errors no matter what kind of mistake I type in my code. Anyone know how to turn it back on?
I've checked my actionscript settings and I have both warning mode and strict mode checked on.
Could it be a socket issue? I admit I have little to no experience working with sockets and only a surface understanding of what that even means. I've added the socket tag. If someone can see that this is clearly nothing to do with sockets, by all means, I'll remove the tag.
UPDATE: 6/22/16
I just reinstalled Adobe Animate CC 2015.2 and no change. I'll try compiling it in flash builder when I get a chance to help pigeon hole the problem. (Edit 6/24: flash builder worked! But my trial version expired the next day so is no longer a viable option).
And I just tried something in the command prompt in Windows 10 as an administrator:
netsh winsock reset catalog
netsh int ip reset reset.log hit
which I found another user on flashdevelop.org used to fix a similar issue, but no change.
And I just tried debugging in Flash Builder. It worked fine (debugger caught bugs) but my trial version expired the next day.
update 6/24
I've tried launching debugger for AIR from within the Adobe Animate CC IDE and it works fine if there are no bugs; it fails to do anything visible (no Iphone emulator, no swf window) if I put a typo or error in the code.
I also just deleted winsock and winsock2, rebooted Windows, then reinstalled winsock and winsock2. No change.
update 6/25
Just tried a system restore in Windows to set all my files and settings and drivers etc. to the way it was a week and a half ago... Also completely uninstalled all Adobe products and reinstalled. No change. I can only imagine that wiping my hard drive and reinstalling Windows would do the trick, but come on, it hasn't come to that has it?
As VC.One suggested, I checked the compile error window (cannot believe I didn't check before! Maybe when I checked before there were no compile errors... Only runtime errors?) and the errors are showing up there. Does this mean that it's catching compile errors but just not runtime errors?
When you force/test a runtime error... make sure you check
Compiler Errors (ALT+F2) and also Output (F2). By Output I mean the window that shows traces (and runtime issues). One of those two should have some feedback for you.
A possible solution is to save a new Workspace. So with those windows for Compiler Errors and Output both open (or tabbed, I tab them next to my Timeline) go to Window (in top options like File/Edit/Debug etc) and choose Workspace, then into that you choose New Workspace. Give it a name in the pop-up and okay everything.
Flash should always load that current workspace (next time, go to Workspace option again, check that your specified workspace [by name] has a tick next to it...).
Possible pitfalls:
1 - Use the debug player
2 - Make sure there isn't somewhere a try/catch enclosing the portion of code that triggers the event that may lead to an error
3 - Socket issue: may be exported in debug mode but swf cannot connect to debugger (it waits in a blank state for 60 seconds I guess)
4 - Does it compile ok? If there are compilation errors you may get an SWF anyway but then it will not start
...
I was using IE10 on Windows 7 but it is such a buggy piece of you-know-what (hanging, crashing, etc.) that I gave up and installed IE11. It has not hung or crashed since installing. But I hate the F12 developer tools! Okay, not completely - there are some very cool new features. What I don't like is that they seem to have dropped several features that I really liked! Unless I'm just missing something... I've searched and searched on Google and Microsoft but all of the help I've found only describes the new features. Here is what I'm missing: Color picker, Ruler, and most of all, the Clear browser cache for this domain. They allege to have a Clear browser cache function, but it doesn't work! So when I make changes to my website, in particular changing graphics, the only way I can see the change is to delete all my temporary files from IE. Then I lose all of my cookies e.g. for automatically logging in to Stack Overflow! IE version 11.0.9600.16428 on Windows 7 Ultimate 64-bit.
The color picker is still available, go to the DOM Explorer tab and there is an eye dropper on the tool bar at the top towards the left. That activates the color picker.
For the issue of serving cached files try toggling on "Always refresh from server" on the network tool (3rd option from the left). This should get you the latest changes from your server.
In my VC++ application I have an embedded browser (MSHTML). It works fine and handles the mouse properly (for instance, clicks and selects are processed OK). However, mouse wheel rotations over the embedded browser do not have any effect. This is my problem.
I am not very familiar with the internals of MSHTML embedding, and OLE in general. This is an wxWidgets application (wxWidgets is a C++ GUI library), and I am making use of its IEHTMLWin component (which hosts an MSHTML control and wraps it in the wxWindow interface). However, I do have the source and am willing to do some debugging.
Forgetting wxWidgets and speaking purely about OLE and MSHTML, what's the right place to start looking for the problem cause? I tried naive googling for variants of "mshtml mouse events" or "mshtml wheel", but didn't give any good pointers.
Should you want to take a look at the code of IEHTMLWin, it can be browsed here. The iehtmlwin.c file (1,5 k lines) has all the OLE-related code and implements all interfaces needed to host a web browser control. It's worth noting that mouse events don't proceed to the containing wxWindow at all (OnMouse is never called).
{UPD} mshtml version: 6.00.2900.3314. Other applications that host this control (including IE) support the wheel. {/UPD}
jdigital hint (regarding winspector) was very helpful. After some message sniffing, I realize that the problem is focus-related. A click on the browser control somehow does not set focus on the browser control (unlike, say, RichEdit), and WM_MOUSEWHEEL is not sent there. So the new problem is setting the focus.
Try Winspector (http://www.windows-spy.com/) which will allow you to see the windows messages. Make sure that the scroll wheel events are getting passed through.