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

From: <bugzilla-noreply_at_freebsd.org>
Date: Thu, 29 Aug 2024 23:20:07 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=280846

--- Comment #4 from Henrich Hartzer <henrichhartzer@tuta.io> ---
Sorry for the wait! Thank you for all of the helpful info and things to
try/check.

I'm using ZFS without swap. I use encrypted rootfs as the installer does it. I
think with geli? -- I wonder if this is partly to blame?

I'm using a SSD.

real memory  = 17179869184 (16384 MB)
avail memory = 16441286656 (15679 MB)

My swapinfo -m is blank.

No tmpfs.

Video is i915kms. I've had this problem on a Thinkpad x200 and now on a x230.

System will usually start to act up before the hang, but the last one it did (a
few minutes ago) had pretty much no warning at all.

I started running top and gstat with those flags and will report back. I might
try to reproduce it, we'll see.

Wayland + Sway these days, but I had it with X11 + DWM.

It might take a few HomeDepot tabs to do it, but it's not totally consistent.
I've tried to reproduce it before and have had no luck.

vm.pfault_oom_attempts: 3
vm.pfault_oom_wait: 10

Here's some current stats:

244 processes: 2 running, 240 sleeping, 2 waiting
CPU:  1.9% user,  0.0% nice,  1.2% system,  0.4% interrupt, 96.6% idle
Mem: 919M Active, 1015M Inact, 134M Laundry, 1606M Wired, 56K Buf, 351M Free
ARC: 573M Total, 456M MFU, 75M MRU, 12K Anon, 6000K Header, 35M Other
     459M Compressed, 567M Uncompressed, 1.23:1 Ratio

dT: 1.003s  w: 1.000s
 L(q)  ops/s    r/s     kB   kBps   ms/r    w/s     kB   kBps   ms/w    d/s    
kB   kBps   ms/d    o/s   ms/o   %busy Name
    0      0      0      0      0    0.0      0      0      0    0.0      0    
 0      0    0.0      0    0.0    0.0| ada0
    0      0      0      0      0    0.0      0      0      0    0.0      0    
 0      0    0.0      0    0.0    0.0| ada1


Please let me know if I missed anything pertinet. I'll hopefully remember to do
a verbose boot.

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