Date: Tue, 28 Jan 2020 10:50:41 +0000 From: bugzilla-noreply@freebsd.org To: net@FreeBSD.org Subject: [Bug 243392] if_vmx(4): Input buffer corruption Message-ID: <bug-243392-7501-IOlQaMAn9D@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-243392-7501@https.bugs.freebsd.org/bugzilla/> References: <bug-243392-7501@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D243392 --- Comment #7 from Andriy Gapon <avg@FreeBSD.org> --- When I run tcpdump on the other end, there is no effect on the speed. There are some potentially interesting snippets in the capture. Here is one where 10.25.200.199 is the vmx side and 10.180.0.51 is the other host that sends data to the vmx host: 115 02:16:28.787281 IP 10.180.0.51.17739 > 10.25.200.199.22: Flags [.], = seq 81774:83160, ack 1, win 1026, options [nop,nop,TS val 3595053000 ecr 1184721692], length 1386 116 02:16:28.787284 IP 10.180.0.51.17739 > 10.25.200.199.22: Flags [.], = seq 83160:84546, ack 1, win 1026, options [nop,nop,TS val 3595053000 ecr 1184721692], length 1386 117 02:16:28.787304 IP 10.180.0.51.17739 > 10.25.200.199.22: Flags [.], = seq 84546:85932, ack 1, win 1026, options [nop,nop,TS val 3595053000 ecr 1184721692], length 1386 118 02:16:28.787307 IP 10.180.0.51.17739 > 10.25.200.199.22: Flags [.], = seq 85932:87318, ack 1, win 1026, options [nop,nop,TS val 3595053000 ecr 1184721692], length 1386 119 02:16:28.883545 IP 10.25.200.199.22 > 10.180.0.51.17739: Flags [.], = ack 83160, win 2192, options [nop,nop,TS val 1184721792 ecr 3595053000], length= 0 120 02:16:28.883576 IP 10.180.0.51.17739 > 10.25.200.199.22: Flags [.], = seq 87318:88704, ack 1, win 1026, options [nop,nop,TS val 3595053097 ecr 1184721792], length 1386 121 02:16:28.887035 IP 10.25.200.199.22 > 10.180.0.51.17739: Flags [.], = ack 83160, win 2192, options [nop,nop,TS val 1184721792 ecr 3595053000,nop,nop,= sack 1 {87318:88704}], length 0 122 02:16:29.336381 IP 10.180.0.51.17739 > 10.25.200.199.22: Flags [.], = seq 83160:84546, ack 1, win 1026, options [nop,nop,TS val 3595053549 ecr 1184721792], length 1386 My reading of the above. Packets 115-118: data packets Packet 119: ack for packet 115 Packet 120: another data packet Packet 121: appears to be exact duplicate of the packet 119 (ack) Packet 122: looks like a retransmission of packet 116 after a noticeable de= lay (0.44s) Well, on another look, packets 119 and 121 are not exactly the same. 121 has this "sack 1 {87318:88704}" that 119 does not have. And it seems like packet 121 is what "confuses" the sending side as the pau= se and retransmission. Not a TCP expert, so don't know if anything is wrong here. --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-243392-7501-IOlQaMAn9D>