Brand new Debian 11 install. Running actual Oracle Virtualbox. Cosmology has this same problem. I have three brand new installs on my BOINC machines. Debian 11, Ubuntu 20.04 LTS, and the latest Solus. Only on Debian does this problem exist. For Cosmology I had to hack app_config.xml in the project directory to make it have the following.

<app_config>
    <app>
        <name>camb_boinc2docker</name>
        <max_concurrent>1</max_concurrent>
    </app>
    <app_version>
        <app_name>camb_boinc2docker</app_name>
        <plan_class>vbox64_mt</plan_class>
        <avg_ncpus>2</avg_ncpus>
    </app_version>
</app_config>

The same problem appears to be happening with Rosetta and Debian 11. Basically the "fail" is being caused by the communication with VirtualBox demanding all 4 CPU. It can never get all 4.

It appears you have the same developer, or they both cut & pasted code from the same place without knowing what it was actually doing. Can someone please provide the proper app_config.xml to make this problem go away?[/img]

Rosetta Python is single threaded so I have no idea why yours are asking for 4 CPUs.
I can't really help but I hope someone can.


Do note however that the setting <max_concurrent>1</max_concurrent> has some kind of bug that actually causes BOINC to download a bunch of work (I've seen reports of hundreds of tasks).

---