[PATCH net v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days

[PATCH net v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days

[Fernando Fernandez Mancera]

This is a recommendation from RFC 8981 and it was intended to be changed
by commit 969c54646af0 ("ipv6: Implement draft-ietf-6man-rfc4941bis")
but it only changed the sysctl documentation.

Fixes: 969c54646af0 ("ipv6: Implement draft-ietf-6man-rfc4941bis")
Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
---
v2: included a comment indicating the value comes from RFC 8981 section
3.8
---
 include/net/addrconf.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 78e8b877fb25..9e96776945e5 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -8,7 +8,8 @@
 
 #define MIN_VALID_LIFETIME		(2*3600)	/* 2 hours */
 
-#define TEMP_VALID_LIFETIME		(7*86400)       /* 1 week */
+/* TEMP_VALID_LIFETIME default value as specified in RFC 8981 3.8 */
+#define TEMP_VALID_LIFETIME		(2*86400)       /* 2 days */
 #define TEMP_PREFERRED_LIFETIME		(86400)         /* 24 hours */
 #define REGEN_MIN_ADVANCE		(2)             /* 2 seconds */
 #define REGEN_MAX_RETRY			(3)
-- 
2.53.0

Re: [PATCH net v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days

[Eric Dumazet]

On Sat, Feb 14, 2026 at 6:25 PM Fernando Fernandez Mancera
<fmancera@suse.de> wrote:
>
> This is a recommendation from RFC 8981 and it was intended to be changed
> by commit 969c54646af0 ("ipv6: Implement draft-ietf-6man-rfc4941bis")
> but it only changed the sysctl documentation.
>
> Fixes: 969c54646af0 ("ipv6: Implement draft-ietf-6man-rfc4941bis")
> Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
> ---

Reviewed-by: Eric Dumazet <edumazet@google.com>

Thanks !

Re: [PATCH net v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to netdev/net.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Sat, 14 Feb 2026 18:25:43 +0100 you wrote:
> This is a recommendation from RFC 8981 and it was intended to be changed
> by commit 969c54646af0 ("ipv6: Implement draft-ietf-6man-rfc4941bis")
> but it only changed the sysctl documentation.
> 
> Fixes: 969c54646af0 ("ipv6: Implement draft-ietf-6man-rfc4941bis")
> Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
> 
> [...]

Here is the summary with links:
  - [net,v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days
    https://git.kernel.org/netdev/net/c/9e371b0ba7f5

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

Re: [PATCH net v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days

[Fernando Fernandez Mancera]

On 2/18/26 2:20 AM, patchwork-bot+netdevbpf@kernel.org wrote:
> Hello:
> 
> This patch was applied to netdev/net.git (main)
> by Jakub Kicinski <kuba@kernel.org>:
> 
> On Sat, 14 Feb 2026 18:25:43 +0100 you wrote:
>> This is a recommendation from RFC 8981 and it was intended to be changed
>> by commit 969c54646af0 ("ipv6: Implement draft-ietf-6man-rfc4941bis")
>> but it only changed the sysctl documentation.
>>
>> Fixes: 969c54646af0 ("ipv6: Implement draft-ietf-6man-rfc4941bis")
>> Signed-off-by: Fernando Fernandez Mancera <fmancera@suse.de>
>>
>> [...]
> 
> Here is the summary with links:
>    - [net,v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days
>      https://git.kernel.org/netdev/net/c/9e371b0ba7f5
> 

FTR; the patch was merged without the fixes tag. I think it would be 
good to backport this to stable kernels because otherwise the 
documentation and the real configured value are contradicting each other.

> You are awesome, thank you!

Re: [PATCH net v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days

[Jakub Kicinski]

On Fri, 20 Feb 2026 16:39:27 +0100 Fernando Fernandez Mancera wrote:
> > Here is the summary with links:
> >    - [net,v2] ipv6: addrconf: reduce default temp_valid_lft to 2 days
> >      https://git.kernel.org/netdev/net/c/9e371b0ba7f5
> >   
> 
> FTR; the patch was merged without the fixes tag. I think it would be 
> good to backport this to stable kernels because otherwise the 
> documentation and the real configured value are contradicting each other.

Yes, because it's a functional change. We can take it in for 7.0 
but IMO the justification is too weak to treat it as a stable fix.
Imagine this from the perspective of someone who was promised 
a _stable_ long term support kernel. If this change breaks their
expectations - what will be our defense? "But we said in the doc 
(that nobody reads) that it was supposed to be 2 days"? Dunno.
Deciding backports is quite tricky, borderline impossible.
I took the executive decision to take the patch ASAP but not to stable.

Of course the stable auto-selection process may suck this patch in,
anyway. But I won't be culpable :)