- 1. Re: Regression in pmtud lvs-nat with ipv6 since at least 3.10 (score: 1)
- Author: Art -kwaak- van Breemen <ard@xxxxxxxxxxxxxxx>
- Date: Tue, 18 Feb 2014 14:37:26 +0100
- Hi Julian, I've made a test setup. With attached patch (agains 3.13.3, but probably generic to >= 3.10) I get this: Feb 18 14:24:41 c43236 kernel: [ 7103.570868] IPVS: Enter: ip_vs_out, net/netfilter
- /html/lvs-devel/2014-02/msg00013.html (13,707 bytes)
- 2. Re: Regression in pmtud lvs-nat with ipv6 since at least 3.10 (score: 1)
- Author: Art -kwaak- van Breemen <ard@xxxxxxxxxxxxxxx>
- Date: Tue, 18 Feb 2014 11:05:33 +0100
- Hi Julian, I was thinking along that line too, since sizeof(struct icmp6hdr) would exactly skip that part, hence pointing ciph at the second half of the first ip address. I try to make a test setup,
- /html/lvs-devel/2014-02/msg00012.html (9,527 bytes)
- 3. Re: Regression in pmtud lvs-nat with ipv6 since at least 3.10 (score: 1)
- Author: Julian Anastasov <ja@xxxxxx>
- Date: Mon, 17 Feb 2014 23:34:43 +0200 (EET)
- Hello, Can you try to change the IPPROTO_ICMPV6 argument of ipv6_find_hdr with -1, eg: ipv6_find_hdr(skb, &icmp_offset, IPPROTO_ICMPV6, &fragoffs, NULL); to: ipv6_find_hdr(skb, &icmp_offset, -1, &fra
- /html/lvs-devel/2014-02/msg00011.html (11,922 bytes)
- 4. Re: Regression in pmtud lvs-nat with ipv6 since at least 3.10 (score: 1)
- Author: Art -kwaak- van Breemen <ard@xxxxxxxxxxxxxxx>
- Date: Mon, 17 Feb 2014 17:16:01 +0100
- Hi, What I think happened is that the ipv6 headers themselves are correctly handled, but that the ipv6 address in the icmp part is written to the wrong place. (See: net/netfilter/ipvs/ip_vs_core.c: i
- /html/lvs-devel/2014-02/msg00010.html (10,679 bytes)
- 5. Regression in pmtud lvs-nat with ipv6 since at least 3.10 (score: 1)
- Author: Art -kwaak- van Breemen <ard@xxxxxxxxxxxxxxx>
- Date: Mon, 17 Feb 2014 16:32:06 +0100
- Hi, I just discovered a major regression in path-mtu-discovery with lvs-nat-ipv6 It seems that at least since 3.10.27, verifiedn on 3.12.10, with the latest known working at least 3.7.10. The problem
- /html/lvs-devel/2014-02/msg00009.html (10,927 bytes)
This search system is powered by
Namazu