[Bug 280846] Low memory freezes / OOM: a thread waited too long to allocate a page

From: <bugzilla-noreply_at_freebsd.org>
Date: Fri, 16 Aug 2024 06:50:43 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=280846

--- Comment #2 from Mark Millard <marklmi26-fbsd@yahoo.com> ---
(In reply to Mark Millard from comment #1)

What is the output like is analogous to in:

# sysctl vm.pfault_oom_attempts vm.pfault_oom_wait
vm.pfault_oom_attempts: 3
vm.pfault_oom_wait: 10

I'll note that:

# sysctl vm.pfault_oom_attempts=-1
vm.pfault_oom_attempts: 3 -> -1

prevents executing the sequence:

        printf("vm_fault_allocate_oom: proc %d (%s) failed to alloc page on
fault, starting OOM\n",
                curproc->p_pid, curproc->p_comm);

        vm_pageout_oom(VM_OOM_MEM_PF);

but it does so by being willing to deadlock/livelock
the system instead. Also other forms of OOM activity
could still potentially happen and such a change
might make that more likely.

-- 
You are receiving this mail because:
You are the assignee for the bug.