Under memory pressure, flushing caches - Zombie AppVM

User Support
1

Hello,

I had an AppVM closing all its windows.
The Zombi-VM remained active, and non-responsive.
… ( terrible joke, right ?! :joy: )

Is this a case of lack of memory allocation?
If so, shouldn’t it have used swap?
shouldn’t QuBes warn the users about this situations?

For new users this type of errors may happen more often.
the experience is not clarified by any alerts.
just a heads-up that maybe a warning message could help users.

thanks,


[2025-11-09 13:12:27] [199573.020750] oom-kill:constraint=CONSTRAINT_NONE,nodemask=(null),cpuset=/,mems_allowed=0,global_oom,task_memcg=/user.slice/user-1000.slice/session-c3.scope,task=Isolated Web Co,pid=48252,uid=1000 
[2025-11-09 13:12:27] [199573.020845] Out of memory: Killed process 48252 (Isolated Web Co) total-vm:4100660kB, anon-rss:837512kB, file-rss:2280kB, shmem-rss:1524kB, UID:1000 pgtables:3344kB oom_score_adj:100
[2025-11-09 13:12:29] [199574.833947] systemd-journald[301]: Under memory pressure, flushing caches.
2

Would you please provide more data:

  1. Qubes r4.3 or r4.3 (and which rc if the latter).
  2. Kernel LTS (default) or Latest (for both dom0 and the AppVM).
  3. Was it a Whonix AppVM or ordinary. And Debian or Fedora if the latter.
3

hi @alimirjamali

.
Host:

  • QuBes → OS 4.2
  • Xen → 4.17.5
  • kernel → 6.12.42-1.qubes.fc37.x86_64 #1 SMP PREEMPT_DYNAMIC

.
AppVM

  • Debian 12 xfce
  • kernel → (default) 6.12.42-1.qubes.fc37
4

How much was the initial memory/max memory of the qube (in qube settings → advanced tab).

cc. @marmarek (is this similar to what I have been experiencing?)

5

here is the info guys - @alimirjamali @marmarek

original ram = 2GB
current ram = 4GB

there was no problems since changed.
but, I wonder if this should not be notified to users when it crashes.