NetBSD-Bugs archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
kern/57676: Document sysctl net.inet6.ip6.gifpmtu for gif(4)
>Number: 57676
>Category: kern
>Synopsis: Document sysctl net.inet6.ip6.gifpmtu for gif(4)
>Confidential: no
>Severity: non-critical
>Priority: low
>Responsible: kern-bug-people
>State: open
>Class: doc-bug
>Submitter-Id: net
>Arrival-Date: Wed Oct 25 19:20:01 +0000 2023
>Originator: Anthony Mallet
>Release: -current
>Organization:
>Environment:
NetBSD cactus 10.99.7 NetBSD 10.99.7 (CACTUS) #11: Wed Oct 4 22:23:03 CEST 2023 troot@cactus:/usr/obj/sys/arch/amd64/compile/CACTUS amd64
>Description:
I stuggled a bit to figure out why my egress gif(4) packets would always be fragmented to what seemed to be a 1280 MTU, even though I had configured a 1460 one.
The attached doc patch would have saved me a bit of UTSL. I guess it's worth mentionning that gif(4) does not do PMTU discovery but instead uses
net.inet6.ip6.gifpmtu to either always use the minimum 1280 MTU or trust the one configured.
Please feel free to rephrase or rewrite my proposal.
Best,
>How-To-Repeat:
>Fix:
Index: share/man/man4/gif.4
===================================================================
RCS file: /cvsroot/src/share/man/man4/gif.4,v
retrieving revision 1.34
diff -u -r1.34 gif.4
--- share/man/man4/gif.4 14 Aug 2018 06:27:44 -0000 1.34
+++ share/man/man4/gif.4 25 Oct 2023 19:11:49 -0000
@@ -140,6 +140,12 @@
When the inner packet is IPv6, the protocol field of the outer packet
is 41
.Dv ( IPPROTO_IPV6 ) .
+.Ss Path MTU discovery
+When the outer packet is IPv6, path MTU discovery is not done. Instead, the
+net.inet6.ip6.gifpmtu
+.Xr sysctl 7
+variable controls the MTU used for egress packets. When set to 0, the minimum
+MTU (1280) is always used. When set to 1, the interface MTU is trusted.
.Ss Security
Malicious party may try to circumvent security filters by using
tunneled packets.
Home |
Main Index |
Thread Index |
Old Index