Heyo
I know it can be a bit daunting xD
I basically had to compile everything locally to see the changes. It's a headache and I don't recommend it if you're not familiar with setting up dev environments.
The simple way for you to try this fix out would be to get the nightly build here: binary-factory.kde.org/job/KStars_Nightly_macos/
Sadly the nightly builds are broken and it's stuck on a build previous to the one that introduced this fix =/
IIRC there was a topic here that talked about fixing this but there seems to have been no change yet :c
Read More...
Me on the other hand bring news of success. I was finally able to do a polar align and nothing crashed!! Now the only weird issue is that the mount seems to slew incorrectly but that's probably something related to config.
To try and help you with your plight.
I'm using ZSH. Using that config allowed me to run both the arm and x86 versions of brew. When compiling kstars I also used an iTerm instance that was using rosetta. So x86_64 all the way through.
When prompted I used Qt5
I had some issues trying to the script to use the correct python because I also use pyenv ocasionally in my dev work and it was giving me problems. But after a while I got it to work.
Had to comment out the lines that copy the Craft ini files.
While it is compiling the indi server it also fails with a missing header. But after installing the missing lib it goes through.
And that's all I think
Read More...
Now it doesn't crash anymore and instead says
2024-01-23 12:17:22.979573+0000 kstars[54517:806952] [org.kde.kstars.ekos.align] "No Target - Please pick an object."Read More...
Yeah I had a bunch of problems too with this. Basically it's trying to use the python with arm64 executable which was installed by brew which also differentiates between arm64 and x86_64.
I ended up running this setup stackoverflow.com/a/68443301.
Read More...
Necroing this post because my post is basically a repost of this. I'm on 3.6.8 and this is still crashing like this. Was the fix rolled out with the version?
My post -> www.indilib.org/forum/ekos/14181-kstars-...ng-align-module.html
Read More...
That said and since you also have an M1 Sonoma. Are you able to boot up ekos with internal indi server with a mount simulator and a ccd simulator and use the align module? For me it crashes almost always. Capture and Solve, load and slew and polar align all crash.
Read More...
I did it for x86_64 instead with rosetta. I don't think it's buildable yet to arm64. It'd be a headache to get indi to build in arm64 and then kstars after.
Read More...
So after quite a few headaches I finally managed to build Kstars and run it with lldb. Funnily enough the first time I ran it it didn't crash (it happens once in a blue moon)
The second time crashed straight away.
This is the backtrace:
(lldb) bt
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x18)
* frame #0: 0x000000010fc42257 libindiclient.2.0.5.dylib`INDI::AbstractBaseClient::getBLOBMode(char const*, char const*) [inlined] std::__1::unique_ptr<INDI::AbstractBaseClientPrivate, std::__1::default_delete<INDI::AbstractBaseClientPrivate>>::get[abi:v160006](this=0x0000000000000018) const at unique_ptr.h:276:105 [opt]
frame #1: 0x000000010fc42257 libindiclient.2.0.5.dylib`INDI::AbstractBaseClient::getBLOBMode(char const*, char const*) [inlined] std::__1::unique_ptr<INDI::AbstractBaseClientPrivate, std::__1::default_delete<INDI::AbstractBaseClientPrivate>>::element_type* getPtrHelper<std::__1::unique_ptr<INDI::AbstractBaseClientPrivate, std::__1::default_delete<INDI::AbstractBaseClientPrivate>>>(p=nullptr) at indimacros.h:133:14 [opt]
frame #2: 0x000000010fc42257 libindiclient.2.0.5.dylib`INDI::AbstractBaseClient::getBLOBMode(char const*, char const*) [inlined] INDI::AbstractBaseClient::d_func(this=0x0000000000000010) at abstractbaseclient.h:39:9 [opt]
frame #3: 0x000000010fc42257 libindiclient.2.0.5.dylib`INDI::AbstractBaseClient::getBLOBMode(this=0x0000000000000010, dev="CCD Simulator", prop="CCD1") at abstractbaseclient.cpp:449:5 [opt]
frame #4: 0x00000001004a22f2 kstars`___lldb_unnamed_symbol13189 + 1026
frame #5: 0x0000000112f8242d QtCore`void doActivate<false>(QObject*, int, void**) + 1133
frame #6: 0x0000000111242330 QtWidgets`QAbstractButtonPrivate::emitClicked() + 112
frame #7: 0x000000011124219a QtWidgets`QAbstractButtonPrivate::click() + 250
frame #8: 0x000000011124347f QtWidgets`QAbstractButton::mouseReleaseEvent(QMouseEvent*) + 271
frame #9: 0x000000011119aa72 QtWidgets`QWidget::event(QEvent*) + 514
frame #10: 0x00000001112e8480 QtWidgets`QPushButton::event(QEvent*) + 80
frame #11: 0x000000011115f7e0 QtWidgets`QApplicationPrivate::notify_helper(QObject*, QEvent*) + 256
frame #12: 0x0000000111162641 QtWidgets`QApplication::notify(QObject*, QEvent*) + 7201
frame #13: 0x0000000112f54779 QtCore`QCoreApplication::notifyInternal2(QObject*, QEvent*) + 169
frame #14: 0x000000011116019e QtWidgets`QApplicationPrivate::sendMouseEvent(QWidget*, QMouseEvent*, QWidget*, QWidget*, QWidget**, QPointer<QWidget>&, bool, bool) + 846
frame #15: 0x00000001111b8030 QtWidgets`QWidgetWindow::handleMouseEvent(QMouseEvent*) + 2048
frame #16: 0x00000001111b6c2b QtWidgets`QWidgetWindow::event(QEvent*) + 203
frame #17: 0x000000011115f7e0 QtWidgets`QApplicationPrivate::notify_helper(QObject*, QEvent*) + 256
frame #18: 0x0000000111160c04 QtWidgets`QApplication::notify(QObject*, QEvent*) + 484
frame #19: 0x0000000112f54779 QtCore`QCoreApplication::notifyInternal2(QObject*, QEvent*) + 169
frame #20: 0x0000000111f6845d QtGui`QGuiApplicationPrivate::processMouseEvent(QWindowSystemInterfacePrivate::MouseEvent*) + 3069
frame #21: 0x0000000111f4e1fb QtGui`QWindowSystemInterface::sendWindowSystemEvents(QFlags<QEventLoop::ProcessEventsFlag>) + 235
frame #22: 0x0000000115ccd4f8 libqcocoa.dylib`QCocoaEventDispatcherPrivate::postedEventsSourceCallback(void*) + 40
frame #23: 0x00007ff8065437d6 CoreFoundation`__CFRUNLOOP_IS_CALLING_OUT_TO_A_SOURCE0_PERFORM_FUNCTION__ + 17
frame #24: 0x00007ff806543779 CoreFoundation`__CFRunLoopDoSource0 + 157
frame #25: 0x00007ff806543548 CoreFoundation`__CFRunLoopDoSources0 + 215
frame #26: 0x00007ff8065421b8 CoreFoundation`__CFRunLoopRun + 919
frame #27: 0x00007ff806541859 CoreFoundation`CFRunLoopRunSpecific + 557
frame #28: 0x00007ff8112df9d9 HIToolbox`RunCurrentEventLoopInMode + 292
frame #29: 0x00007ff8112df616 HIToolbox`ReceiveNextEventCommon + 201
frame #30: 0x00007ff8112df531 HIToolbox`_BlockUntilNextEventMatchingListInModeWithFilter + 66
frame #31: 0x00007ff809ab17b9 AppKit`_DPSNextEvent + 880
frame #32: 0x00007ff80a3a9f64 AppKit`-[NSApplication(NSEventRouting) _nextEventMatchingEventMask:untilDate:inMode:dequeue:] + 1304
frame #33: 0x00007ff809aa2d32 AppKit`-[NSApplication run] + 603
frame #34: 0x0000000115ccc174 libqcocoa.dylib`QCocoaEventDispatcher::processEvents(QFlags<QEventLoop::ProcessEventsFlag>) + 2132
frame #35: 0x0000000112f51296 QtCore`QEventLoop::exec(QFlags<QEventLoop::ProcessEventsFlag>) + 502
frame #36: 0x0000000112f54d13 QtCore`QCoreApplication::exec() + 131
frame #37: 0x0000000100016d46 kstars`___lldb_unnamed_symbol4191 + 22870
frame #38: 0x00000002011c6386 dyld`start + 1942This is a consistent crash in every mac with sonoma I tried (I don't have access to a mac without sonoma installed)
Does anyone have any insight on what might be causing this?
Read More...
Since the last post I spent an ungodly amount of time trying to build this locally with no luck. I wanted to build a debug version to which I could attach an lldb session to get a backtrace but no luck.
I've tried running kstars on multiple macs with all of them crashing. Is there anything else I can do? Any way to get a more descriptive error?
Read More...
I'll just add some info.
I'm running 3.6.8 stable version.
I've added a video of the whole process of launching and crashing. It says that there's an indi server already running because of the previous crash. If I start this after a laptop restart the indi server starts without the warning. But it still crashes.
Read More...