On Fri, 13.11.15 14:49, Lennart Poettering (lenn...@poettering.net) wrote: > Heya! > > So, I am tempted to make the following changes to systemd, and was > wondering about opinions about it:
[...] > Of course, it also has potential to break some services, but I think > defaults like 1K and 8K are high enough to make this the exception, > not the rule. In summary, I think we gain more by improving security > and robustness through putting strict limits on everything we do, than > we lose. > > Users could of course unset these defaults, to lift the limits. And > packages could lift the limit in their unit files too, if they know > that they are too low for their specific service. So, I implemented all of this now. I accidentally pushed this directly into master unfortunately. Sorry for that! Was supposed to become a PR, but I was on the wrong branch. Anyway, fort hose who do reviews: please do a post-commit review. Anyway, for now I settled for the default TasksMax= setting of 512 for all units, plus 4096 for the per-user slices and 8192 for each nspawn instance. Let's see how this will work out. Sorry again for the direct push. Lennart -- Lennart Poettering, Red Hat _______________________________________________ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel
