I cut down the timeline and removed 45 mins of my time line, still fcpx crashed.
I used to run on Mojave and my machine is constantly crashing, took it apple store they changed the logic board. When I try to run the same time line it was crashing and one of the Genius Bar technican told me it is due to power, so I have remove the cable from surge protector and plugged into the wall. Now I don't have crashes due to power but I am getting various crashes.
I removed the audio plugin which was the culprit for the crashes and I made sure that there is no clip in my timeline refers to the plugin removed which was the culprit.
Now I reduced the timeline, removed the plugins, still I am getting crashed but this time the messages are different. Any help in guiding to identify the issue will be greatly appreciated
Audio Plug -- Accusonus ERA5
I will attach the core dump in the attachment section.
Thread 32 crashed due to memory heap corruption. That doesn't mean damaged RAM. An application uses memory in four main categories: code, data, stack and heap. The heap is the portion that is dynamically allocated. To try and protect from programming errors the heap is fenced by "guard values". If those are changed, then some thread within the app did that inappropriately.
Since any thread within an app has full access to all of the app's memory, any plugin thread can crash the entire app. The exception is some brand new plugins that have moved to the FxPlug 4 framework which run in a separate host process. You could ask why did they design it that way, but the debate over in-process vs out-of-process extensions is an old one in computer science that dates back many decades.
I don't see any "usual suspects" of poorly-behaved plugins but the crash dump was converted to a .PDF which wrapped lines and is difficult to read.
I suggest you make a duplicate of the timeline, open that duplicate timeline, then remove all effects by CMD+A to select all clips, and Edit>Remove Effects to remove those. Then do operations on that clean timeline with no effects and see if it crashes.
If it does not, then repeat the above procedure but remove effects on 1/2 of the timeline, then the other 1/2, then 1/4, etc, until you find what region is implicated.
Unfortunately some plugins and effects will be in FCP memory even if they are not used on any clip. So the above procedure is not guaranteed to find the problem, but it is a common troubleshooting method.
Following are my comments
I have removed the audio plugin altogether from the system and made sure that the effects selection are removed from the clips. Also I have created a new user and removed all the plugins from the timeline. I will try your suggestions and post it here.
Looks like to me, fcpx issue but I want to rule hardware issue because I just got the logic board replaced and cleanuped my Mac and installed Big Sur.
I will upload the dump in a text file for your reference.
Please attach any new crash logs so I can compare them.
Are you using Apple or 3rd-party RAM? We just had a case where we spent a huge amount of time trying to debug a crash which was later traced to 3rd-party RAM.
Can you describe what you are doing when it crashes? Is it the same user action or is anything in common in the user actions? Thread 31 implies there was a Compressor operation in progress, and thread 30 says a background sharing operation was happening.
Do you have FCP background rendering enabled? Try disabling that in FCP preferences>General>Playback. Then delete any render files by File>Delete Generated Library Files>Delete Render Files>All. Try the operation without any render files.
Deleting render files does not delete all cached items such as analysis files, optical flow files, waveforms, etc. To delete those, select the library in FCP and in Inspector use Storage Locations>Modify Settings>Cache and place the cache in a separate folder (ideally on your fastest drive). Subsequently you can safely delete the .fcpcache bundle in that folder. That is in case there might be a corrupt generated file.
You have installed FxFactory, so make sure that is updated to the latest version. If it still crashes, use the below instructions to completely uninstall FxFactory (that is just a test, not permanently): fxfactory.com/support/
I am using a third party RAM not an apple RAM, my RAM is 32 GB not sure about the Vendor.
3.7GHz 6-core 9th-generation Intel Core i5 processor, Turbo Boost up to 4.6GHz8GB 2666MHz DDR4 memoryRadeon Pro 580X with 8GB of GDDR5 memory2TB Fusion Drive storage
I am not sure why the background rendering is happening during import, I normally uncheck that every time before I export the timeline. Also I delete all the render files from the project before exporting. I have not cleared the cache files which I can give a try. I have got 7 to 8 panic alert and I gave up and took the computer to Apple to look at it why it is unstable. During export or after the export failure if I use my browser the system crash.
I followed your direction and removed all the effects so there is no effect including a color correction in the timeline. After that, my system crashed a couple of times. I also got an export failure i.e, it is not able to write to file not sure how to debug that issue. I have around 920 GB of space free in the local hard drive and the output file is only 42 GB. I tried the same in the external drive also, the same result.
I have removed half of the timeline and got at least two panics and I wasted almost two days where I cant able to export a single time due to which I took it back to the Apple store to figure out why the export is making the system so unstable and any operation after that results in system crash.
I will not get the computer for another two to three days and hopefully, I can try the cache suggestion you have made.
I have not installed any fx factory for the new user and I haven't removed it either. As I said before I removed all the text labels and audio plugin accusonus ERA5 which I thought was the culprit for the export failures but I am unable to confirm it.
Do you think another vendor's memory is an issue? I can get a set of Apple memory if needed.
I have created a new destination for 4K from the compressor and I expose it so that when I try to share my timeline I use the 4K destination that I have created using Compressor. I always use the same destination while sharing which is a 3840x2160 59.94 frames at the rate of 35K.
Thank you so much for your time and appreciate all your inputs.
....During export or after the export failure if I use my browser the system crash...I have removed half of the timeline and got at least two panics and I wasted almost two days where I cant able to export a single time due to which I took it back to the Apple store to figure out why the export is making the system so unstable and any operation after that results in system crash....I will not get the computer for another two to three days and hopefully, I can try the cache suggestion you have made....Do you think another vendor's memory is an issue? I can get a set of Apple memory if needed.....
If the entire system ever has a kernel panic or totally hangs, that is not an FCP problem and cannot be fixed at the FCP level. Re 3rd-party RAM, a lot of people use that with success. However in this case, we spent a huge amount of time going through crash logs, chasing plugins, trying various things at the FCP level, only to positively identify the cause as 3rd-party RAM. Fortunately the OP posted the resolution which enabled making that determination: www.fcp.co/forum/4-final-cut-pro-x-fcpx/...-kernel-crash#115454
That does not mean in your case it is RAM, but your problem is apparently not caused by FCP, rather FCP is the victim of a lower-level system instability. Temporarily replacing the 3rd-party RAM with Apple RAM would be a reasonable troubleshooting step, given the symptoms. The problem could be hardware, config or kernel-mode drivers. You can run this terminal command to inspect what 3rd-party kernel-mode drivers are loaded:
kmutil inspect | grep -v com.apple
Re hardware, the Genius Bar has access to much more rigorous "bench diagnostics" which if needed can be run overnight or over multiple days. They may resist doing this unless you insist. I had a 2015 iMac 27 which was spontaneously rebooting and it only failed diagnostics when they were run overnight by the Genius Bar.
In the past I have used either Memtest for Mac or Rember (GUI Memtest) for testing RAM. You can set # of times each RAM chip is tested / repeats or let test process run for awhile. This has been a good way to stress test RAM as sometimes it can be intermittent hence overnight runs may cause crash. Repeatedly hitting the RAM chips should cause bad memory to surface.
Generally this gets your Mac running and fans blazing.
Not sure if it works in new M1, but should work on i5 Intel Macs.
I have given the computer to apple and after 3 days, they acknowledged memory issues when the CPU is maxing out which is strange,. I haven't seen that personally but they're able to reproduce and they're reaching out to their engineering team for malloc issues. In the meanwhile, they're ordering apple memory chips and want to rule out that it is not an OWC memory issue.. Will keep updating this thread based on the progress.. Thanks for all the inputs, I will validate all the inputs once I get the computer back.
I got a call from Apple stating that the OWC Memory modules are the problem. THey have tried using Apple memory modules and they can reproduce shutdown/crashes. I spoke to OWC about that and they're graceful enough to replace the memory module after two years. I will keep the update posted as I get new memory modules and how the export works after the memory module replacement.
I have sent my old OWC memory for replacement and got an additional 32Gb OWC RAM from Amazon yesterday. I am able to export 4K 150 min timeline without any challenge. I am trying to add the audio plugin and primary color correction and export the timeline one more to see whether the issues have been resolved. Meanwhile, the computer is more stable and there are no crashes so far, looking promising but I will confirm in few days. Appreciate all the feedback, very useful to move in the right direction. Apple support provided great support which made my life easier by identifying the memory issues, ran various test for multiple days etc.
Hey thanks for posting your solution. This is good for the community!
RAM can indeed cause KPs. OWC has always been good about backing up their products. I had a bad stick that was also intermittent. OWC replaced with no problems.
Happy editing, Greg
I want to second Greg in thanking you for sharing the solution. Performance issues can be very hard to troubleshoot and it looks like the advice given by Joe and Greg has pointed you in the right direction. This thread will be helpful to the entire community. Good job!
Latest Update, I have done multiple exports with no failure. The feedback is on target which helped me to work with Apple support to get the issues resolved. As Ron said, hope the discussion can help someone in the future who has similar symptoms. Thanks, everyone, and thoroughly enjoyed the expertise and the support
Thanks for posting your updates! I'm experiencing near identical issues, but combined with some weird screen issues despite all hardware passing all tests.
How many sticks of RAM did you end up replacing? At least one of mine has just failed and I'm awaiting replacement from OWC. FCPX crashing constantly while I'm using some borrowed RAM, 1 stick is OWC the others are Apple. (long story)
Sorry, I just saw your post, I have replaced both OWC 32 GB(16GB each) and they sent me a replacement stick. I added two more sticks bumping to 64 GB now and I have no issues. The main issues are 1) Logic board 2) memory sticks. If memory sticks are not resolving it then it might be something with logic board. Please take it to Apple and ask them to validate it. I hope this info helps you.