I had the same segmentation fault 11 problem. I tried everything. I just downloaded the BackEnd 0.27.4 DMG package. In there there is a MythFillDatabase. It ran perfectly.
I was previously using the 0.27.3 version.
JLB
MythFillDatabase segmentation fault
Re: MythFillDatabase segmentation fault
When I tried using "Mythbackend" Mac OSX, I had nothing but trouble and segfaults all the time.
Solution for me is to run "Mythbackend" using Mythbuntu inside VMWare Fusion on the Mac OSX. Now my Mac Mini system is rock solid with no crashes. I HIGHLY recommend running "Mythbackend" in VMWare Fusion on the Mac OSX. (if you can afford to get a copy of VMWare Fusion)
ps: My experience has been that "Mythfrontend" has always worked fine on the Mac OSX (using it for many years now with my backend running on Linux)
Solution for me is to run "Mythbackend" using Mythbuntu inside VMWare Fusion on the Mac OSX. Now my Mac Mini system is rock solid with no crashes. I HIGHLY recommend running "Mythbackend" in VMWare Fusion on the Mac OSX. (if you can afford to get a copy of VMWare Fusion)
ps: My experience has been that "Mythfrontend" has always worked fine on the Mac OSX (using it for many years now with my backend running on Linux)
Re: MythFillDatabase segmentation fault
Whatever works for you...but many of us have been running the backend on OS X for years without crashes.firmdog wrote:When I tried using "Mythbackend" Mac OSX, I had nothing but trouble and segfaults all the time.
Solution for me is to run "Mythbackend" using Mythbuntu inside VMWare Fusion on the Mac OSX. Now my Mac Mini system is rock solid with no crashes. I HIGHLY recommend running "Mythbackend" in VMWare Fusion on the Mac OSX. (if you can afford to get a copy of VMWare Fusion)
ps: My experience has been that "Mythfrontend" has always worked fine on the Mac OSX (using it for many years now with my backend running on Linux)
Craig
Formerly the MacPorts guy.
Re: MythFillDatabase segmentation fault
The fact that it's reporting a crash with QString::isEmpty() indicates a bug in QT on OSX. QStrings are default initialised and yet it's crashing with 'Bad Access' which is suggestive of it trying to access an unreserved or uninitialised area of memory.
I've no experience of debugging on OSX, but the information reported in the backtrace isn't pointing at a MythTV bug and without further information I can't be of more help. I suggest you look at what version of QT is being used and what versions of OSX since it may be a bad interaction between the two.
I've no experience of debugging on OSX, but the information reported in the backtrace isn't pointing at a MythTV bug and without further information I can't be of more help. I suggest you look at what version of QT is being used and what versions of OSX since it may be a bad interaction between the two.
Re: MythFillDatabase segmentation fault
MacPorts provides version 4.8.6 and quite a few ports depend on it. There are a number of cases on the MacPorts bug tracker but none of them seem to cause crashes related to QStrings:stuartm wrote:The fact that it's reporting a crash with QString::isEmpty() indicates a bug in QT on OSX. QStrings are default initialised and yet it's crashing with 'Bad Access' which is suggestive of it trying to access an unreserved or uninitialised area of memory.
I've no experience of debugging on OSX, but the information reported in the backtrace isn't pointing at a MythTV bug and without further information I can't be of more help. I suggest you look at what version of QT is being used and what versions of OSX since it may be a bad interaction between the two.
https://trac.macports.org/query?status= ... r=priority
Also, it seems odd that this bug apparently affects a few OS X users but many of us never see it.
Craig
Formerly the MacPorts guy.
Re: MythFillDatabase segmentation fault
Well that's why I suggested the exact version of OS X may also be related.pvr4me wrote:Also, it seems odd that this bug apparently affects a few OS X users but many of us never see it.
It's nearly impossible to crash on a QString unless it's being accessed from another thread to the one in which it was created. I'm not very familiar with the Schedules Direct parser, but the rest of mythfilldatabase is single threaded, and I'd done recall seeing any threading used in the DD/SD code.