rtsold: pass sending router address to other and managed script

[franco_opnsense.org]

Summary

Partially based on an old pfSense modification that never made it
to FreeBSD upstream.

See: https://github.com/fichtner/pfsense-tools/commit/2a78e6483

Test Plan

Running for many years on pfSense and OPNsense

[markj]

usr.sbin/rtsold/rtsol.c

83

It looks like ntopbuf is not initialized at the points where rtsol_input() invokes the scripts.

[franco_opnsense.org]

usr.sbin/rtsold/rtsol.c

83

It's being set via line 289 inside warnmsg() unconditionally. Scripts are called in line 307 and 324. I don't mind making it more explicit if you think that would help.

[markj]

usr.sbin/rtsold/rtsol.c

83

Sorry, I see. Yes, I think making it a separate statement would be preferable. For good measure, I would make sure that ntopbuf is zeroed at the beginning of the function.

[franco_opnsense.org]

now avoids raw access to ntopbuf from scripts call

[franco_opnsense.org]

usr.sbin/rtsold/rtsol.c

83

Created access through new variable via inet_ntop() call which mimics the other invokes. I am not sure initializing ntopbuf does anything since only inet_ntop() and snprintf() are being used to write and check for success. If an error occurred any one of those invokes that could taint the data for the next use anyway.

[markj]

I think this is ok. Is there any reason not to supply the sender's addr to the other script types, if only for consistency?

usr.sbin/rtsold/rtsol.c

83

I was mostly concerned about the possibility of someone in the future shuffling things around such that ntopbuf is left uninitialized, but your approach is better.

[franco_opnsense.org]

The only other script is the DNS script and it looks like -u already does append the sender address to that script's data.

I am not a FreeBSD committer... how will the two rtsold patches move into the tree? Thanks for your help.

[markj]

In D31518#710875, @franco_opnsense.org wrote:

The only other script is the DNS script and it looks like -u already does append the sender address to that script's data.

I see, sorry again for the misapprehension.

I am not a FreeBSD committer... how will the two rtsold patches move into the tree? Thanks for your help.

I will commit them this weekend if that works for you.

[franco_opnsense.org]

That would be great, thanks!