<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Trex ARP strange issue in Tools</title>
    <link>https://community.cisco.com/t5/tools/trex-arp-strange-issue/m-p/4187648#M579</link>
    <description>&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;This is my setup to benchmark my server&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;A href="https://ibb.co/0KWwT8Z" target="_blank" rel="noopener nofollow ugc"&gt;https://ibb.co/0KWwT8Z&lt;/A&gt;&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;Trex version v2.88&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;Trex server MAC&lt;/P&gt;&lt;PRE&gt;ens2f0: f8:f2:1e:5c:bf:78
ens2f1: f8:f2:1e:5c:bf:79&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;DUT server MAC&lt;/P&gt;&lt;PRE&gt;eth0: fa:16:3e:1c:be:82
eth1: fa:16:3e:d5:37:b1&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;trex_cfg.yaml&lt;/P&gt;&lt;PRE&gt;# cat /etc/trex_cfg.yaml
### Config file generated by dpdk_setup_ports.py ###

- version: 2
  #interfaces: ['07:00.0', '07:00.1']
  interfaces: ['ens2f0', 'ens2f1']
  port_info:
      - ip: 10.66.0.100
        default_gw: 10.66.1.33
      - ip: 10.68.0.100
        default_gw: 10.68.3.217

  platform:
      master_thread_id: 0
      latency_thread_id: 8
      dual_if:
        - socket: 0
          threads: [1,2,3,4,5,6,7,16,17,18,19,20,21,22,23]&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;My scenario file&lt;/P&gt;&lt;PRE&gt;- duration : 5
  generator :
          distribution : "seq"
          clients_start : "16.0.0.1"
          clients_end   : "16.0.0.255"
          servers_start : "48.0.0.1"
          servers_end   : "48.0.255.255"
          clients_per_gb : 10
          min_clients    : 10
          dual_port_mask : "1.0.0.0"
          tcp_aging      : 0
          udp_aging      : 0
  cap_info :
     - name: cap2/rtp_160k.pcap
       cps : 10.0
       ipg : 10000
       rtt : 10000
       w   : 4
     - name: cap2/rtp_250k_rtp_only_1.pcap
       cps : 10.0
       ipg : 10000
       rtt : 10000
       w   : 4
     - name: cap2/rtp_250k_rtp_only_2.pcap
       cps : 10.0
       ipg : 10000
       rtt : 10000
       w   : 4 &lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;Run command&lt;/P&gt;&lt;PRE&gt;./t-rex-64 --cfg /etc/trex_cfg.yaml -f /tmp/sfr.yaml -d 10 -m 1 -c 1 -k 2 --nc&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;In Test i am seeing Trex sending UDP stream but RX end not receiving any data and result is 100% packet drops. when i run tcpdump on my DUT device i can see its sending UDP stream to Trex RX end using correct MAC address. fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;DUT server tcpdump&lt;/P&gt;&lt;PRE&gt;    16.0.0.17.58005 &amp;gt; 48.0.0.17.1024: [no cksum] UDP, length 1231
15:38:34.237250 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype IPv4 (0x0800), length 378: (tos 0x0, ttl 125, id 11822, offset 0, flags [none], proto UDP (17), length 364)
    16.0.0.18.9873 &amp;gt; 48.0.0.18.1026: [no cksum] UDP, length 336
15:38:34.237260 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype IPv4 (0x0800), length 1492: (tos 0x0, ttl 125, id 11852, offset 0, flags [none], proto UDP (17), length 1478)
    16.0.0.15.23195 &amp;gt; 48.0.0.15.1026: [no cksum] UDP, length 1450
15:38:34.237266 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype IPv4 (0x0800), length 434: (tos 0x0, ttl 125, id 11832, offset 0, flags [none], proto UDP (17), length 420)&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;As soon as test start i am seeing correct ARP entry mac-ip but in 2 second it get&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;STRONG&gt;incomplete&lt;/STRONG&gt;&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;even my test is running and keep sending data.&lt;/P&gt;&lt;PRE&gt;[root@foo-2 ~]# arp -a | grep 10.68.0.100
? (10.68.0.100) at f8:f2:1e:5c:bf:79 [ether] on eth1
[root@foo-2 ~]# arp -a | grep 10.68.0.100
? (10.68.0.100) at &amp;lt;incomplete&amp;gt; on eth1&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;Any idea why ARP getting incomplete in 2 second?&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;&amp;nbsp;&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;&lt;STRONG&gt;UPDATE:&lt;/STRONG&gt;&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;When i run tcpdump on DUT server then i found Trex RX side interface ask for ARP and DUI replied back but when DUT ask for ask then Trex not replying back (&lt;A href="https://10.68.3.217/" target="_blank" rel="noopener nofollow ugc"&gt;10.68.3.217&lt;/A&gt;&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;is eth1 IP of DUT server)&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;who is 01:03:05:07:09:00 and 01:03:05:07:09:01 ?&lt;/P&gt;&lt;PRE&gt;[root@foo-2 ~]# tcpdump -i eth1 -nn arp -e -vv
tcpdump: listening on eth1, link-type EN10MB (Ethernet), capture size 262144 bytes
16:03:26.527165 f8:f2:1e:5c:bf:79 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), length 60: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 (01:03:05:07:09:00) tell 10.68.0.100, length 46
16:03:26.527192 f8:f2:1e:5c:bf:79 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), length 60: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.3.217 (01:03:05:07:09:01) tell 10.68.0.100, length 46
16:03:26.527201 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Reply 10.68.3.217 is-at fa:16:3e:d5:37:b1, length 28
16:03:32.107249 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:33.131245 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:34.155245 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:35.183759 fa:16:3e:d5:37:b1 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:36.203260 fa:16:3e:d5:37:b1 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:37.227240 fa:16:3e:d5:37:b1 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), lengt&amp;nbsp;&lt;/PRE&gt;&lt;P&gt;T&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Mon, 23 Nov 2020 18:55:46 GMT</pubDate>
    <dc:creator>satish.txt1</dc:creator>
    <dc:date>2020-11-23T18:55:46Z</dc:date>
    <item>
      <title>Trex ARP strange issue</title>
      <link>https://community.cisco.com/t5/tools/trex-arp-strange-issue/m-p/4187648#M579</link>
      <description>&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;This is my setup to benchmark my server&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;A href="https://ibb.co/0KWwT8Z" target="_blank" rel="noopener nofollow ugc"&gt;https://ibb.co/0KWwT8Z&lt;/A&gt;&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;Trex version v2.88&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;Trex server MAC&lt;/P&gt;&lt;PRE&gt;ens2f0: f8:f2:1e:5c:bf:78
ens2f1: f8:f2:1e:5c:bf:79&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;DUT server MAC&lt;/P&gt;&lt;PRE&gt;eth0: fa:16:3e:1c:be:82
eth1: fa:16:3e:d5:37:b1&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;trex_cfg.yaml&lt;/P&gt;&lt;PRE&gt;# cat /etc/trex_cfg.yaml
### Config file generated by dpdk_setup_ports.py ###

- version: 2
  #interfaces: ['07:00.0', '07:00.1']
  interfaces: ['ens2f0', 'ens2f1']
  port_info:
      - ip: 10.66.0.100
        default_gw: 10.66.1.33
      - ip: 10.68.0.100
        default_gw: 10.68.3.217

  platform:
      master_thread_id: 0
      latency_thread_id: 8
      dual_if:
        - socket: 0
          threads: [1,2,3,4,5,6,7,16,17,18,19,20,21,22,23]&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;My scenario file&lt;/P&gt;&lt;PRE&gt;- duration : 5
  generator :
          distribution : "seq"
          clients_start : "16.0.0.1"
          clients_end   : "16.0.0.255"
          servers_start : "48.0.0.1"
          servers_end   : "48.0.255.255"
          clients_per_gb : 10
          min_clients    : 10
          dual_port_mask : "1.0.0.0"
          tcp_aging      : 0
          udp_aging      : 0
  cap_info :
     - name: cap2/rtp_160k.pcap
       cps : 10.0
       ipg : 10000
       rtt : 10000
       w   : 4
     - name: cap2/rtp_250k_rtp_only_1.pcap
       cps : 10.0
       ipg : 10000
       rtt : 10000
       w   : 4
     - name: cap2/rtp_250k_rtp_only_2.pcap
       cps : 10.0
       ipg : 10000
       rtt : 10000
       w   : 4 &lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;Run command&lt;/P&gt;&lt;PRE&gt;./t-rex-64 --cfg /etc/trex_cfg.yaml -f /tmp/sfr.yaml -d 10 -m 1 -c 1 -k 2 --nc&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;In Test i am seeing Trex sending UDP stream but RX end not receiving any data and result is 100% packet drops. when i run tcpdump on my DUT device i can see its sending UDP stream to Trex RX end using correct MAC address. fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;DUT server tcpdump&lt;/P&gt;&lt;PRE&gt;    16.0.0.17.58005 &amp;gt; 48.0.0.17.1024: [no cksum] UDP, length 1231
15:38:34.237250 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype IPv4 (0x0800), length 378: (tos 0x0, ttl 125, id 11822, offset 0, flags [none], proto UDP (17), length 364)
    16.0.0.18.9873 &amp;gt; 48.0.0.18.1026: [no cksum] UDP, length 336
15:38:34.237260 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype IPv4 (0x0800), length 1492: (tos 0x0, ttl 125, id 11852, offset 0, flags [none], proto UDP (17), length 1478)
    16.0.0.15.23195 &amp;gt; 48.0.0.15.1026: [no cksum] UDP, length 1450
15:38:34.237266 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype IPv4 (0x0800), length 434: (tos 0x0, ttl 125, id 11832, offset 0, flags [none], proto UDP (17), length 420)&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;As soon as test start i am seeing correct ARP entry mac-ip but in 2 second it get&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;STRONG&gt;incomplete&lt;/STRONG&gt;&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;even my test is running and keep sending data.&lt;/P&gt;&lt;PRE&gt;[root@foo-2 ~]# arp -a | grep 10.68.0.100
? (10.68.0.100) at f8:f2:1e:5c:bf:79 [ether] on eth1
[root@foo-2 ~]# arp -a | grep 10.68.0.100
? (10.68.0.100) at &amp;lt;incomplete&amp;gt; on eth1&lt;/PRE&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;Any idea why ARP getting incomplete in 2 second?&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;&amp;nbsp;&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;&lt;STRONG&gt;UPDATE:&lt;/STRONG&gt;&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;When i run tcpdump on DUT server then i found Trex RX side interface ask for ARP and DUI replied back but when DUT ask for ask then Trex not replying back (&lt;A href="https://10.68.3.217/" target="_blank" rel="noopener nofollow ugc"&gt;10.68.3.217&lt;/A&gt;&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;is eth1 IP of DUT server)&lt;/P&gt;&lt;P class="_1qeIAgB0cPwnLhDF9XSiJM"&gt;who is 01:03:05:07:09:00 and 01:03:05:07:09:01 ?&lt;/P&gt;&lt;PRE&gt;[root@foo-2 ~]# tcpdump -i eth1 -nn arp -e -vv
tcpdump: listening on eth1, link-type EN10MB (Ethernet), capture size 262144 bytes
16:03:26.527165 f8:f2:1e:5c:bf:79 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), length 60: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 (01:03:05:07:09:00) tell 10.68.0.100, length 46
16:03:26.527192 f8:f2:1e:5c:bf:79 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), length 60: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.3.217 (01:03:05:07:09:01) tell 10.68.0.100, length 46
16:03:26.527201 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Reply 10.68.3.217 is-at fa:16:3e:d5:37:b1, length 28
16:03:32.107249 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:33.131245 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:34.155245 fa:16:3e:d5:37:b1 &amp;gt; f8:f2:1e:5c:bf:79, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:35.183759 fa:16:3e:d5:37:b1 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:36.203260 fa:16:3e:d5:37:b1 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), length 42: Ethernet (len 6), IPv4 (len 4), Request who-has 10.68.0.100 tell 10.68.3.217, length 28
16:03:37.227240 fa:16:3e:d5:37:b1 &amp;gt; ff:ff:ff:ff:ff:ff, ethertype ARP (0x0806), lengt&amp;nbsp;&lt;/PRE&gt;&lt;P&gt;T&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 23 Nov 2020 18:55:46 GMT</pubDate>
      <guid>https://community.cisco.com/t5/tools/trex-arp-strange-issue/m-p/4187648#M579</guid>
      <dc:creator>satish.txt1</dc:creator>
      <dc:date>2020-11-23T18:55:46Z</dc:date>
    </item>
  </channel>
</rss>

