Page MenuHomeFreeBSD

Add Parameter Problem type for Incomplete Header Chain
AcceptedPublic

Authored by thj on Aug 7 2019, 2:02 PM.
Tags
None
Referenced Files
Unknown Object (File)
Sun, Apr 7, 12:43 PM
Unknown Object (File)
Sun, Apr 7, 12:31 PM
Unknown Object (File)
Sun, Apr 7, 12:26 PM
Unknown Object (File)
Sun, Apr 7, 11:43 AM
Unknown Object (File)
Feb 20 2024, 1:22 AM
Unknown Object (File)
Dec 20 2023, 12:47 AM
Unknown Object (File)
Dec 10 2023, 7:55 PM
Unknown Object (File)
Nov 10 2023, 12:09 PM
Subscribers

Details

Reviewers
jtl
bz
gallatin
Group Reviewers
network
Summary

Split out from https://reviews.freebsd.org/D16851

Add Parameter Problem type for Incomplete Header Chain

Add Parameter Problem message type for "IPv6 First Fragment has
incomplete IPv6 Header Chain" as described in RFC7112.

Add a stat counter to track the occurances of this message and teach
netstat to report these values.

Diff Detail

Lint
Lint Passed
Unit
No Test Coverage
Build Status
Buildable 25729
Build 24311: arc lint + arc unit

Event Timeline

bz requested changes to this revision.Nov 19 2019, 2:27 PM
bz added inline comments.
sys/netinet/icmp6.h
591

By adding it to not-the-end you are breaking old netstat binaries (not that they'd be happy anyway as the struct size will change). Normally we'd do append to have a chance to at least properly read the "so far known" counters even if we get more data.

sys/netinet6/in6.h
646

This seems unrelated to adding the value and the stats field?

usr.bin/netstat/inet6.c
1039

Not sure I'd call it bad header chain, given this is something more specific. It really means no-ulp-in-first-frag, right? frag1-no-ulp? or simply bad-frag-hdr-chain? I think it is essential that this is related to the first fragment.

no or incomplete upper layer header in first fragment << that too long a line?

This revision now requires changes to proceed.Nov 19 2019, 2:27 PM
thj marked 3 inline comments as done.Aug 28 2020, 10:06 AM

Updated diff coming I changed the netstat line to "no or incomplete upper layer header in first fragment" which did not wrap until there were 10 figures of packets.

  • Address review comments
This revision is now accepted and ready to land.Aug 28 2020, 11:21 AM