360 Equirectangular Images and Drone Images Causing Stalling

How did you install WebODM (docker, installer, etc.)?

Docker

What's your browser and operating system? (Copy/paste the output of https://www.whatismybrowser.com/)

Browser: Chrome
OS: WebODM exists on an Ubuntu server but is viewed from a Windows PC
Output of whatismybrowser: Chrome 104 on Windows 11

What is the problem?

I have been attempting to process a mixed data set (drone and GoPro 360) of images and have been getting odd errors. Everything else processes just fine but these datasets freeze. Note: The dataset usually processes when I use the drone footage and regular 360 images and pick spherical camera. The data just looks really bad and jumbled. The dataset causing the issue consists of drone images and 360 photos that have been split into their 6 equirectangular components and converted from tifs to jpgs.

They get about 3 hours in (11 hours on more intense settings) and stop processing but the task log does not show any errors.

Every time I refresh, the processing timer, resets to whenever it got stalled I guess. So if I check and it says 3:11:52 and I leave it for 5 minutes, it will say 3:16:52 and appear to be processing.
But if I leave and come back to the page or refresh, the timer resets to 3:11:52.

The processes also don’t successfully cancel. This occurred before and after a WebODM update. I am currently running WebODM 1.9.15

How can we reproduce this? (What steps trigger the problem? What parameters are you using for processing? Include screenshots. If you are having issues processing a dataset, you must include a copy of your dataset uploaded on Dropbox, Google Drive or https://dronedb.app)

I upload the dataset and run it like normal. Everything looks good until anywhere from 3-11 hours in depending on the number of images I include. That is when the counter stops and resets every time I refresh the page but it appears to still be running. The only parameters I have changed are pc quality to ultra and feature quality to ultra. Previously, I have used the DSM+DTM setting and changed other parameters but it didn't change the outcome. See additional details on community.opendronemap.org: https://community.opendronemap.org/t/drone-360-hybrid-imaging-bug/12653

Smallest dataset that failed is here: https://drive.google.com/drive/folders/1bb332I94u-fzltfinUf6bdrcSaRyY3vB?usp=sharing

Dataset that completed after down. update, start command was run is here: https://drive.google.com/drive/folders/1jQDaD9kMYusxCxay0QjEFLSiMcUBGIG4?usp=sharing

Start with the smallest failing dataset linked in the issue and compare it with the dataset that completed after the update and start command. Check the task log and refresh/cancel behavior; done means the mixed drone/equirectangular job completes or fails with an actionable error and can be cancelled.