Thomas Haller

> It feels like something is consuming a large amount of memory. I think this report is not actionable. Needs more information...

this branch would need to rebased to latest `main` branch. Also, the multiple commits should be squashed in one.

Sorry for no reply for such a long time... There are changes like ```diff msgid "Interfaces" -msgstr "" +msgstr "" ``` which seems not a correct translation update. I cannot...

Hi. You are correct. Patch makes sense to me! The branch should be rebased to latest `master` (I don't have permission to do that). Rebasing will result in conflicts, due...

I think this will be fixed by https://github.com/firewalld/firewalld/pull/1187

> Firewalld already drops capabilities/privileges. See commit [fb0532e](https://github.com/firewalld/firewalld/commit/fb0532e8a200b15b7e83077aec380c35c0695475) and [1380196](https://github.com/firewalld/firewalld/commit/13801962073f478c68d818b314091badcf8b5614). Is that desirable to reimplement dropping capabilities, instead of dropping them via systemd? Is the use-case to support non-systemd systems?...

I think upstream should target rather recent systemd versions and enable useful (sandboxing) options. This would then require older distros to patch those options out. The benefit is that downstream...

fixed by https://github.com/firewalld/firewalld/pull/1166

> This can be a problem when used to create a sequence number and that number overflows the uint32_t maximum value when it is incremented. What is exactyly the problem...

> In our code base, we compile the code such that it will abort on overflow, so these devices abort quickly. That seems a questionable thing to do. It is...