systemd-oomd Looks Like It Will Come Together For systemd 247

[phoronix]

Phoronix: systemd-oomd Looks Like It Will Come Together For systemd 247

With systemd 246 releasing soon, it looks like the new out-of-memory daemon will be merged shortly afterwards in allowing sufficient time for testing ahead of systemd 247...

http://www.phoronix.com/scan.php?page=news_item&px=systemd-oomd-coming-soon

[halo9en]

I wonder how it differs from nohang (which supports PSI, low mem warnings, and systemd units).

Edit: so if I'm not mistaken it's just using Facebook's oomd, which is suited for large data centers.

[boxie]

I wonder how it differs from nohang (which supports PSI, low mem warnings, and systemd units).

Edit: so if I'm not mistaken it's just using Facebook's oomd, which is suited for large data centers.

If I am correct on this - Gnome and KDE are both moving towards launching applications in their own CGroups and this oom daemon kills procs in cgroups if enabled.

My guess is that all apps by default in the future with have the flag to allow this oomd to do its job

[Ulmo]

system-doomd?

[Danny3]

Hopefully this can help Firefox too with tab crashes.

[milkylainen]

Systemd-oomd needs to commit kamikaze so it can free resources for important stuff.

[tildearrow]

Why not fix the problem in the kernel?

Like not swap executable code when the system is near-OOM?

[andyprough]

Three words: No thank you.

systemd deciding which of my apps to kill and when to kill them sounds like a disaster waiting to happen. I can monitor my own system memory just fine, thank you very much. There's conky and applets and htop and glances and system monitors and notifiers all over the place I can use.

[xnor]

One word: ignorance.
Again, people that don't know what they're talking about and haven't got the first clue about how Linux handles OOM situations ideologically condemn it because it has "systemd" in its name.

[Shiba]

system-doomd?

systemd-oom-nom-nom