AIJ fails with universalJavaApplicationStub quit unexpectedly

[mattwood]

Hi, All:

I installed AIJ on my work computer (IT controlled).  It ran fine while they were looking over my shoulder, but when I tried to restart it after a reduction, I get an error message "universalJavaApplicationStub quit unexpectedly" and "The application AstroImageJ can't be opened."  I installed again this morning (with IT again remoting to enter credentials), and it worked until it failed again just now.

M1 iMac 2023 model.  Ventura 13.5 OS.

Any advice would be most welcome!  

Cheers,
Matt Wood
Texas A&M Unviersity-Commerce

[karenacollins]

Hi Matt,

Unfortunately, we do not have an M1-based Mac available here for testing. We'll have to think on an approach to troubleshooting this. Are you installing one of the currently posted packages "AstroImageJ_v5.3.3.00-mac-arm_64Bit.dmg" or "AstroImageJ_v5.3.3.00-mac-x86_64Bit.dmg"? If you are installing the mac-arm version, could you try the mac-x86 version to see if it has the same problem? If I understand correctly mac-x86 builds will work on M1 machines, but maybe slower. If nothing else, this might give us an idea if the problem is in the M1 build or is a more general problem with M1 machines. As a long shot, if you have an Intel-based mac in the same environment, it would be helpful to know if it runs solid on it as it does for us.

Karen

[mattwood]

Good morning, Karen:

It does run perfectly on my other machines, and this one is a university-administered one, so it's possible something with those guardrails is causing the problem.  I have to get IT involved to install something, so I think I'll just reduce my data on my laptop or home machine for the time being.

Thanks so very much for your continued development and involvement with AIJ - it's a wonderful environment for reducing time-series photometry.  

I do wish there was a faster plate solve, but I know that's not AIJ's fault (and it's nice that feature is included)!

Cheers,
Matt

[karenacollins]

Hi Matt,

Thank you for the feedback. Can you clarify if any of your non-university-administered machines is an M1-based Mac?

As FYI, we have spent some time brainstorming other ways of implementing plate-solving. Windows users have the option of installing a local version of a solver that AIJ can directly interface with. It is much fast under most circumstances. But unfortunately that is a windows-only package (ansvr). We're looking for something that we can implement directly in Java so that it works seamlessly on all platforms.

If you do find any other hints that the reported problem is specific to the Mac arm platform (or the arm build of AIJ), please do let us know here.

Karen