#getting this on 95% of renders
33 messages · Page 1 of 1 (latest)
lots of users repeated error on every render
@deft cradle is this due to my cdn?
its accessibly publicly and not restricted by cors
sorry to tag, just that all renders with audio on my cdn is failing now
4.0.261 also failing :/
honestly so cooked now
Error: Compositor error: Invalid data found when processing input 0: <remotion::errors::ErrorWithBacktrace as core::convert::From<ffmpeg_next::util:
:Error>>::from 1: remotion::opened_stream::open_stream 2: remotion::commands::execute_command 3: remotion::main 4: std::sys::backtrace::__rust_begin_short_backtrace 5: main 6: __libc_start_call_main 7: __libc_start_main@GLIBC_2.17 8: <unknown> at onMessage (/var/task/index.js:104310:41) at processInput (/var/task/index.js:59397:9) at processInput (/var/task/index.js:59400:9) at processInput (/var/task/index.js:59400:9) at Socket.onData (/var/task/index.js:59419:9) at Socket.emit (node:events:518:28) at addChunk (node:internal/streams/readable:559:12) at readableAddChunkPushByteMode (node:internal/streams/readable:510:3) at Readable.push (node:internal/streams/readable:390:5) at Pipe.onStreamRead
more errors 
genuinely dont understand how this happens randomly haha
🥲 continuous failure and idk how to solve
genuinely tried everything
not sure what to do, its been hours and renders are still not working. literally all of a sudden
it must be something i am doing but i cannot figure it out as its been perfect for ages
at this point im willing to pay for support lmao
been down for about 5 hours, loads of renders erroring :(
how am i the only one having this issue right now, i have tried and deployed functions and sites on so many versions
ALL fail with this error
and i have no idea how to solve it
@deft cradle feel bad to tag you as i know you must be busy
but in real pain here right now 😢
getting this on 95% of renders
@distant igloo it seems like this is just an invalid audio file
you can submit it at remotion.dev/report if you want me to have a look
Hi Jonny
it also happens during just videos too
it also happens to images now, it says it cannot load image
the thing is I did not change anything, all of a sudden it has just started failing
for context, no videos are the exact same, users are generating different voiceovers, adding different background videos, background audios, images etc
so its all different and not just one file that keeps failing
i have no idea how to debug, these files were not invalid before, and are all publicly accessible
i am very confused
also i cannot submit a report because i do not have an output file because the video never renders
For images:
Error loading image with src: https://cdn-clippie.com/sdxl/4979298e-8109-4881-b390-818da51d0e26-cute%20anime-peaceful.webp at cancelRender ...
For videos with just multiple videos in a sequence and nothing else:
Command failed with exit code 1: /var/task/ffprobe -v error -select_streams a:0 -show_entries stream=channels:stream=start_time:format=duration:format=format_name -of default=nw=1 /tmp/remotion-assetsla6kdud6ma/remotion-assets-dir/819626678712666.mp4 [mov,mp4,m4a,3gp,3g2,mj2 @ 0xea305a0] moov atom not found /tmp/remotion-assetsla6kdud6ma/remotion-assets-dir/819626678712666.mp4: Invalid data found when processing input at makeError (/var/task/index.js:77754:17) at handlePromise (/var/task/index.js:78514:33) at process.processTicksAndRejections (node:internal/process/task_queues:95:5) at async getAudioChannelsAndDurationWithoutCache (/var/task/index.js:103671:22) at async getAudioChannelsAndDurationUnlimited (/var/task/index.js:103704:22) at execute ....
I am happy to share logs, as much detail as possible and even the code
I don't know how to find the cause for this
Happy to pay for support if that helps too ^
just really need to get prod back online
@distant igloo just post or send the link of the videos/images you want to embed for us to analyze
indeed looks perfectly fine to me
i think the problem you have is that the machine rendering the images hits a rate limit from cdn-clippie.com, maybe gets a captcha
otherwise ffmpeg would not say moov atom not found because it is there
Hmm, only 14 requests mitigated
or is this not the correct thing i should be looking for?
maybe those ones are it?
i don't think so, we have had over 100 failed renders
i do strongly believe the render failed because from those URLs not the actual assets were served but something else