std: bad_alloc error inside a ubuntu docker container

[aniket@aus.co.in]

I'm trying to run metashape cli on an ubuntu docker container with 68 GB of RAM. I'm not sure if it is an out of memory error or not. I've also added 20 Gb swap also. Even then it fails with the same error. The dataset has 250 images. Looking at the memory usage profiles it doesn't seem to be a out of memory issue. Any ideas for the possible cause of failure.

[Alexey Pasumansky]

Hello aniket,

Can you please provide the part of the log from the node related to the failed operation? If there is any sensible information (like image filenames or project paths) just remove it from the log.

[aniket@aus.co.in]

here are the logs attached

[Alexey Pasumansky]

Hello aniket,

According to the log, you are trying to generate the mesh model in Arbitrary mode from the dense cloud, which is very resource demanding operation in Metashape 1.7 and is not, actually, suitable for the large dense clouds.
In Metashape 1.8 the mesh generation mode for this surface+source data type combination has been modified and shouldn't require excessive amounts of RAM.

[aniket@aus.co.in]

I can't find link to download Metashape 1.8

[Alexey Pasumansky]

Hello aniket,

It's available as pre-release now:
https://www.agisoft.com/forum/index.php?topic=13736.0
The official release is expected in the very end of December.

Alternatively, you may consider building a tiled model instead of mesh, if you need to visualize large area in high resolution.