On premise model tiler - capture output

We are building a process that automates the execution of the on premise tiler Node applications. This process will be running on Linux and we are running a shell process with a bash command to start the Node application. With this method we are able to capture the output of the pointcloud-tiler as all of is output/process messages are written to standard out. This is not the same for the model-tiler. Some of the output is written to standard out, however the main long running step that generates the output like “Generate Tiles [====================] 37/37(100%)” is not available. You can easy replicate this scenario by running the model-tiler from the command line and piping the output to a text file. When you look at the text file you will see that some of the output seen in the terminal is missing.

Can you please suggest how we can monitor the progress of the model-tiler?

Hi - Just so that I understand correctly and can give you the right solution, is your goal to view the progress of the tilers, or are you looking into integrate it into a larger architecture that can then track the progress programmatically, or maybe even something else?

We are looking to integrate the tilers into a larger architecture, where we need to execute the tiler processes programmatically. As part of executing them programmatically we would also like to track their progress programmatically. I hope this answers your question.

You can take a look at the --format-log-json option which is available for all the tiling pipelines. This outputs a JSON which you can use as a parsing mechanism programmatically.

Thanks - that works great :-). Should probably update the tiler pdf documentation to include this option.