<html>
<head>
<base href="https://bugzilla.netfilter.org/">
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - nfqueue randomly drops packets with same tuple"
href="https://bugzilla.netfilter.org/show_bug.cgi?id=1766#c4">Comment # 4</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - nfqueue randomly drops packets with same tuple"
href="https://bugzilla.netfilter.org/show_bug.cgi?id=1766">bug 1766</a>
from <span class="vcard"><a class="email" href="mailto:antonio.ojea.garcia@gmail.com" title="Antonio Ojea <antonio.ojea.garcia@gmail.com>"> <span class="fn">Antonio Ojea</span></a>
</span></b>
<pre>An interesting observation, the problem only seems to happen when at least one
of the DNAT destinations is in the same namespace where the nfqueue program
runs, I imaging this causes the packet to follow a different codepath than if
the packet is sent out.
What I'm puzzled is why the packets gets dnated twice, after __nf_queue and
before nf_reroute
1500 60 10.244.1.3:45957->10.244.1.5:53(udp) __nf_queue
0xffff9523b3207080 16 <empty>:3178406 4026533244 0 veth271ea3e0:5
0x0800 1500 60 10.244.1.3:45957->10.244.1.5:53(udp) skb_ensure_writable
0xffff9523b3207080 16 <empty>:3178406 4026533244 0 veth271ea3e0:5
0x0800 1500 60 10.244.1.3:45957->10.244.1.5:53(udp) skb_ensure_writable
0xffff9523b3207080 16 <empty>:3178406 4026533244 0 veth271ea3e0:5
0x0800 1500 60 10.244.1.3:45957->10.244.1.5:53(udp)
inet_proto_csum_replace4
0xffff9523b3207080 16 <empty>:3178406 4026533244 0 veth271ea3e0:5
0x0800 1500 60 10.244.1.3:45957->10.244.1.5:53(udp)
inet_proto_csum_replace4
0xffff9523b3207080 16 <empty>:3178406 4026533244 0 veth271ea3e0:5
0x0800 1500 60 10.244.1.3:45957->10.244.2.4:53(udp) nf_reroute</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are watching all bug changes.</li>
</ul>
</body>
</html>