[Bug 282673] ipfw tags are lost while transit via if_epair
- In reply to: bugzilla-noreply_a_freebsd.org: "[Bug 282673] ipfw tags are lost while transit via if_epair"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Tue, 12 Nov 2024 17:21:02 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=282673 --- Comment #3 from bugs.freebsd.org@mx.zzux.com --- (In reply to Andrey V. Elsukov from comment #1) Thanks for the link to the explanation of this, because there is no word about mbuf_tag stripping in 'man if_epair'. Though 'man ipfw' says Tags are "sticky", meaning once a tag is applied to a packet by a matching rule it exists until explicit removal. Tags are kept with the packet everywhere within the kernel, but are lost when the packet leaves the kernel. ... Note: since tags are kept with the packet everywhere in kernelspace, they can be set and unset anywhere in the kernel network subsystem (using the mbuf_tags(9) facility), not only by means of the ipfw(4) tag and untag keywords I think that silent unconditional (eg. via sysctl) behaviour is contrary to the ipfw documentation. And by the way, a packet tagged by the host may be also useful inside jail and vice versa. But I need simply fully functional loopback allowing any action at any pass, not first and only. -- You are receiving this mail because: You are the assignee for the bug.