Recently, my LXC container running the github linux runner, failed. The error message was really confusing:

Failed to create CoreCLR, HRESULT: 0x8007054F

I’ve traced the problem down to:

[pid 680] sched_getaffinity(679, 128, 0x7f21afa0ee50) = -1 EINVAL (Invalid argument)

But why should sched_getaffinity() fail? The only reason I found in the manpage is:

   EINVAL (sched_getaffinity() and, in kernels before 2.6.9, sched_setaffinity()) cpusetsize is smaller than the size of the affinity mask used by the
          kernel.

So, the root cause was the new Linux kernel 6.1 running on this machine, which has increased the number of maximum CPUs. A recompiled kernel with lower setting fixed the issue.