I can't help solve the problem, but I can supply some more information. I've just begun to experience the same problem with IDL v8.2 sp2 on an iMac. I've attached machine information and part of the crash report below. I've applied a few updates lately, OS X Update 10.9.3, but the problem was occurring prior to this. I've updated to Java 8 at May 12, 2014, 8:41:53 AM and to XQuartz v2.7.6 at May 19, 2014, 9:02:14 AM, which immediately predates my problems. I don't seem to have the problem when running from the command line, only from the workbench Machine info Model Name: iMac Model Type: 27-inch, Late 2013 Model Identifier: iMac14,2 Board Identifier: Mac-27ADBB7B4CEE8E61 Processor Name: Intel Core i7 CPU 4771 Processor Model: Haswell Processor Speed: 3.5 GHz Number of Processors: 1 Total Number of Cores: 4 Memory Size: 16 GB 1600 MHz DDR3 Graphics Model: NVIDIA GeForce GTX 780M 4096 MB Assembled in: China Manufactured during: November 2013 L1 Cache: 64 KB L2 Cache: 256 KB (per Core) L3 Cache: 8 MB Processor Bitrate: 64 bit EFI Bitrate: 64 bit Kernel Bitrate: 64 bit Direct Media Interface: 5.0 GT/s Boot ROM Version: IM142.0118.B00 System SMC Version: 2.15f2 Power Source: UPS Installed Hardware Universal Unique Identifier: 00471D98-181A-534F-B99A-45F4DB59CEFD Minimum Supported Operating System: OS X 10.8.4 (Build 12E4022) (Mountain Lion) Current Boot Operating System: OS X 10.9.3 Build 13D65 (Mavericks) Maximum Supported Operating System: Latest version of OS X 10.9 (Mavericks) Hardware Overview: Crash Report: Model Name: iMac Process: idlde [20180] Path: /Applications/exelis/*/idlde.darwin.x86_64.app/Contents/MacOS/idlde Identifier: org.eclipse.eclipse Version: 3.5 (3.5) Code Type: X86-64 (Native) Parent Process: launchd [213] Responsible: idlde [20180] User ID: 502 Date/Time: 2014-05-23 11:56:53.577 -0500 OS Version: Mac OS X 10.9.3 (13D65) Report Version: 11 Anonymous UUID: 86DA3AC9-CB89-C059-8F3B-0772D17699DB Crashed Thread: 29 Java: IDL Exception Type: EXC_BAD_ACCESS (SIGSEGV) Exception Codes: KERN_INVALID_ADDRESS at 0x0000000000000050 VM Regions Near 0x50: --> __TEXT 0000000100000000-0000000100004000 [ 16K] r-x/rwx SM=COW /Applications/exelis/*/idlde.darwin.x86_64.app/Contents/MacOS/idlde Application Specific Information: Java information: Exception type: Bus Error (0xa) at pc=11c41c9ce Java VM: Java HotSpot(TM) 64-Bit Server VM (20.65-b04-462 mixed mode macosx-amd64)

Dear Wouter, Could you rename, or delete the following hidden directory located in your home directory?: .idl After deleting it, could you re-start IDL Workbench. Did that help? Cheers, Fernando

Hi Fernando, Thanks for the reply, but it unfortunately deleting ".idl" folder doesn't help. I've tried it before and the problem stayed, I've -completely- reinstalled IDL and the problem stayed, and I just deleted the .idl folder and the problems pops up again when plotting, saying it's a java thingy. If you have any other suggestions, I'm all ears. Wouter

Hi Wouter, Try starting IDLDE as follow: from the DOS prompt: cd C:\Program Files\Exelis\IDL83\bin\bin.x86_64 idlde -outofprocess Does that help? If you start IDL Command Line, does it work? In the case of using a Mac, you can try from the XQuartz terminal window the same thing: idlde -outofprocess Cheers, fernando

Hi Fernando, I spent yesterday plotting throughout the entire day (woo!). The good news is, IDL crashed less often using idlde -outofprocess. The bad news is, IDL still crashed around 10 times yesterday. Interestingly, I've not been able to find any error logs of the crashes. Either they're hidden, or not produced at all. Most crashes occured when trying to save an image, however not always. So, while it's an improvement, I don't think this is how IDL should behave... Any other suggestions? I'll try anything!

Wouter When I encountered the problem I reported, I was in the process of developing a widget application using new graphics. Once I straightened out some problems with that code, the crashes disappeared. I would suggest checking for orphan pointers and objects in your code. Dave

Hi Dave, Thanks for helping out, but I'm not sure the problem is caused by errors in the code. You see, I'm using IDL mainly to analyze and visualize data. In many occasions the code is custom made for a particular data set and I'm manually copy-pasting lines of code in the IDL> command line. So, here's an example of new graphics code, I would copy-paste: c1=contour(datamatrix,rgb_table=ct,/fill,n_levels=10,axis_style=0,dimensions=[800,800],zrange=[min(datamatrix),max(datamatrix)]) xaxis=axis('X',location=0,target=c1,coord_transform=[number1,number2],tickinterval=1.5) yaxis=axis('Y',location=0,target=c1,coord_transform=[number1,number2],tickinterval=1.5) c2=contour(datamatrix,color='black',/overplot,n_levels=10,c_label_show=0) If I copy this a couple of times (e.g. 4/5 times) onto IDL's command line, IDL crashes, guaranteed. If the graphic pops up and I press the 'save' button, there's a 50/50 chance IDL will crash. I've tried messing around with the object naming; same or different names for the objects, but it doesn't make a difference. I've also tried this with hardly any code at all, so that I was sure there were no orphan pointers or objects. And it took a bit longer than usual, but eventually IDL crashed after repetitive plotting and saving a simple line [p = plot(findgen(10))]. This still happens when I clear the memory of the null object that remains after closing the graphics window. I dont't know what it is. IDL says it java. I think it might also be memory related (even though the systems I'm working on have heaps of memory available). Again, I'm open to all suggestions. Cheers, Wouter

Hi Wouter, I have an idea. Try to reproduce the crash, but this time, run IDL inside Dependency Walker. Dependency Walker will show a lot more system information during the crash. The idea is the following, start Dependecy Walker, then click in File->Open. Then, look for the IDL.exe file. After that, the application will show some information. Now, go to Profile --> Start Profiling. IDL Workbench will start. Work as normally, and try to reproduce the problem. See what Dependency Walker shows when IDL crashes. Maybe there will be a clue in that output. Cheers, fernando