17:39:11 
 tests= 
 _IN_VALIDITY_RPBL_BLOCKED=0. 
 autolearn= 
 fc49- 
 org 
 From: tglaser@b1-systems.de 
  
 On Tue, 4 Nov 2025, Daniel Gröber wrote: 
  
 >It falls on you to keep up with system changes then. Hope you're an avid 
 >NEWS reader :-). 
  
 I am. 
  
 >Thanks for the hint on which pseudoheader to use! 
  
 More help on  or, 
 if needed, #debian-init on OFTC. 
  
 >anyhow. My preferred solution is porting the fixes to Trixie too, I'm just 
 >not sure whether the needed changes are "small" enough for a stable update 
 >at this point :D. 
  
 That’s why one can ask the SRM for pre-approval. 
  
 >To get there without hacks or re-architecting ifupdown I see a need to 
 >transition such systems to having `inet6 auto` lines in /e/n/interfaces so 
 >that ifupdown can come in and decide to launch the DHCPv6 client. 
  
 But I want to specifically NOT use a DHCPv6 client, only rtsol. 
  
 >Hmm. I think I know whats going on but can you post output for: 
 > 
 >    $ ps fax | grep dhcpcd   # to see which instance types are active 
 >    $ sysctl net.ipv6.conf | grep -E 'accept_ra |autoconf' 
  
 tg@tgb1:~ $ ps fax | grep dhcpcd 
  3088 ?        S      0:00 dhcpcd: [manager] [ip4] [ip6] 
  3089 ?        S      0:00  \\_ dhcpcd: [privileged proxy] 
  3163 ?        S      0:00  |   \\_ dhcpcd: [BPF ARP] wlan0 192.168.0.208 
  3090 ?        S      0:00  \\_ dhcpcd: [network proxy] 
  3091 ?        S      0:00  \\_ dhcpcd: [control proxy] 
 15069 pts/4    S+     0:00              \\_ grep dhcpcd 
 tg@tgb1:~ $ sysctl net.ipv6.conf | grep -E 'accept_ra |autoconf' 
 E: /usr/bin/mksh: sysctl: inaccessible or not found 
 1|tg@tgb1:~ $ cat /proc/sys/net/ipv6/conf/wlan0/accept_ra 
  
 0 
 tg@tgb1:~ $ cat /proc/sys/net/ipv6/conf/wlan0/autoconf 
  
 0 
  
 This is from the system now running your code. 
  
 On my bullseye systems, which behave as expected, both are 1. 
  
 >I've added more hacks to make this work. See 
 >https://salsa.debian.org/debian/ifupdown/-/merge_re 
 quests/28/diffs?commit_id=acc9455eef698b4d51f08c3dfa87e31bdcc1147e 
 > 
 >To test it you have to either revert to ifupdown=0.8.44 before upgrading or 
 >touch /run/network/dhcpcd-inet6-stanza-hack.needed to get the equivalent 
 >behaviour. LMK how that works out on your machine. 
  
 I can pile on the extra commits and rebuild, but I fail to see 
 how something in /run would persist across reboots… 
  
 >I am still hoping we can show upstream the right path here, but that's an 
 >ongoing effort everyone's welcome to help with on the upstream issue 
 >tracker :-). 
  
 I’m overwhelmed by the stuff already under my wings… 
  
 >here. The only case where things get tricky is inet6 auto with dhcp=1 but I 
 >don't think that's what you're thinking about? 
  
 I’m… not in enough context to answer that, either. 
  
 We’ll see if the latest iteration helps… 
  
 bye, 
 //Thorsten 
 -- 
 Thorsten Glaser 
 Linux / Unix Developer 
 Tel.: +49 160 91168501 
 E-Mail: tglaser@b1-systems.de 
  
 B1 Systems GmbH 
 Osterfeldstraße 7 / 85088 Vohburg / https://www.b1-systems.de/ 
 GF: Ralph Dehner / Unternehmenssitz: Vohburg / AG: Ingolstadt, HRB 3537 
  
 --- SoupGate-Win32 v1.05 
  * Origin: you cannot sedate... all the things you hate (1:229/2)