<?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 High CPU on subnet scan in ExtremeSwitching (VSP/Fabric Engine)</title>
    <link>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120502#M3099</link>
    <description>&lt;P&gt;Hello,&lt;/P&gt;&lt;P&gt;I would like to understand why a network scan (nmap doing a ping scan) going to a L2VSN could cause control plane to exhibit high CPU.&lt;/P&gt;&lt;P&gt;Obviously we see a lot of ARPs here (broadcast) as the firewall interface toward destination LAN tries to resolve members of the scanned subnet (here: /21).&lt;/P&gt;&lt;P&gt;All switches which have an endpoint configured in that i-sid (flex-uni) experience high CPU in bcmINTR process and also messages about CPP Tx queue getting saturated.&lt;/P&gt;&lt;P&gt;We do use arp inspection but the NNI is explicity set to trusted. The fabric switches themselves do not have an IP interface in that VLAN, it's all L2VSN with flex-uni.&lt;/P&gt;&lt;P&gt;Why is this kind of traffic affecting the control plane so much? (tried on 5520 running FE 8.10.3.0 and 9.1.3.0)&lt;/P&gt;&lt;P&gt;Thx&lt;/P&gt;&lt;LI-CODE lang="markup"&gt;1 2025-10-10T22:55:20.893+02:00 XXX CP1 - 0x0002474c - 00000000 GlobalRouter CPU INFO CPP: 60 percent of fbufs are in use: 1829 in Tx queue, 13 in RxQueue0 0 in RxQueue1 0 in RxQueue2 0 in RxQueue3 0 in RxQueue4 0 in RxQueue5 0 in RxQueue6 0 in RxQueue7  Pkts in Q0 (type): 13(10)

...

Sat Oct 11 2025 02:17:22.785013 1 cpp.c :6081[lcy-te][1154-1742]cbcp-main.x:cppProcRxFrame :CPP: cppProcRxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 IP OR ARP len=56 arrived from 04-01:02:03:04:05:06 updated_port=TX-NNI vid=0x19d pid=2 code=0000 qos=0 PEH 0x019d0000, PktProcCode 0

Sat Oct 11 2025 02:17:22.785074 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/1 pid=2
Sat Oct 11 2025 02:17:22.785120 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/2 pid=2
Sat Oct 11 2025 02:17:22.785192 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/3 pid=2
Sat Oct 11 2025 02:17:22.785229 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/4 pid=2
Sat Oct 11 2025 02:17:22.785265 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/6 pid=2&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Mon, 13 Oct 2025 17:11:43 GMT</pubDate>
    <dc:creator>jeronimo</dc:creator>
    <dc:date>2025-10-13T17:11:43Z</dc:date>
    <item>
      <title>High CPU on subnet scan</title>
      <link>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120502#M3099</link>
      <description>&lt;P&gt;Hello,&lt;/P&gt;&lt;P&gt;I would like to understand why a network scan (nmap doing a ping scan) going to a L2VSN could cause control plane to exhibit high CPU.&lt;/P&gt;&lt;P&gt;Obviously we see a lot of ARPs here (broadcast) as the firewall interface toward destination LAN tries to resolve members of the scanned subnet (here: /21).&lt;/P&gt;&lt;P&gt;All switches which have an endpoint configured in that i-sid (flex-uni) experience high CPU in bcmINTR process and also messages about CPP Tx queue getting saturated.&lt;/P&gt;&lt;P&gt;We do use arp inspection but the NNI is explicity set to trusted. The fabric switches themselves do not have an IP interface in that VLAN, it's all L2VSN with flex-uni.&lt;/P&gt;&lt;P&gt;Why is this kind of traffic affecting the control plane so much? (tried on 5520 running FE 8.10.3.0 and 9.1.3.0)&lt;/P&gt;&lt;P&gt;Thx&lt;/P&gt;&lt;LI-CODE lang="markup"&gt;1 2025-10-10T22:55:20.893+02:00 XXX CP1 - 0x0002474c - 00000000 GlobalRouter CPU INFO CPP: 60 percent of fbufs are in use: 1829 in Tx queue, 13 in RxQueue0 0 in RxQueue1 0 in RxQueue2 0 in RxQueue3 0 in RxQueue4 0 in RxQueue5 0 in RxQueue6 0 in RxQueue7  Pkts in Q0 (type): 13(10)

...

Sat Oct 11 2025 02:17:22.785013 1 cpp.c :6081[lcy-te][1154-1742]cbcp-main.x:cppProcRxFrame :CPP: cppProcRxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 IP OR ARP len=56 arrived from 04-01:02:03:04:05:06 updated_port=TX-NNI vid=0x19d pid=2 code=0000 qos=0 PEH 0x019d0000, PktProcCode 0

Sat Oct 11 2025 02:17:22.785074 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/1 pid=2
Sat Oct 11 2025 02:17:22.785120 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/2 pid=2
Sat Oct 11 2025 02:17:22.785192 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/3 pid=2
Sat Oct 11 2025 02:17:22.785229 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/4 pid=2
Sat Oct 11 2025 02:17:22.785265 1 cpp.c :1595[lcy-ve][1154-1742]cbcp-main.x:cppTxFrame :CPP: cppTxFrame: dst=ff-ff-ff-ff-ff-ff src=00-09-0f-09-02-1e typ=0806 port 1/6 pid=2&lt;/LI-CODE&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Mon, 13 Oct 2025 17:11:43 GMT</pubDate>
      <guid>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120502#M3099</guid>
      <dc:creator>jeronimo</dc:creator>
      <dc:date>2025-10-13T17:11:43Z</dc:date>
    </item>
    <item>
      <title>Re: High CPU on subnet scan</title>
      <link>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120509#M3100</link>
      <description>&lt;P&gt;Please open a ticket so this can be analyzed.&lt;/P&gt;&lt;P&gt;Thanks Roger&lt;/P&gt;</description>
      <pubDate>Tue, 14 Oct 2025 15:47:32 GMT</pubDate>
      <guid>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120509#M3100</guid>
      <dc:creator>Roger_Lapuh</dc:creator>
      <dc:date>2025-10-14T15:47:32Z</dc:date>
    </item>
    <item>
      <title>Re: High CPU on subnet scan</title>
      <link>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120510#M3101</link>
      <description>&lt;P&gt;Does that mean it is not normal and should be treated in hardware and not hitting CPU?&lt;/P&gt;</description>
      <pubDate>Tue, 14 Oct 2025 18:18:34 GMT</pubDate>
      <guid>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120510#M3101</guid>
      <dc:creator>jeronimo</dc:creator>
      <dc:date>2025-10-14T18:18:34Z</dc:date>
    </item>
    <item>
      <title>Re: High CPU on subnet scan</title>
      <link>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120517#M3102</link>
      <description>&lt;P&gt;No it is not normal that those ARPs should hit the CPU. This issue is&amp;nbsp;CFD-13329 and is fixed in 9.3.0.&lt;/P&gt;</description>
      <pubDate>Wed, 15 Oct 2025 08:16:24 GMT</pubDate>
      <guid>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120517#M3102</guid>
      <dc:creator>Ludovico_Steven</dc:creator>
      <dc:date>2025-10-15T08:16:24Z</dc:date>
    </item>
    <item>
      <title>Re: High CPU on subnet scan</title>
      <link>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120519#M3103</link>
      <description>&lt;P&gt;Fix was also back ported to 9.2.1.0 (&lt;A class="" href="https://jira.extremenetworks.com/browse/VOSS-33244" target="_blank" rel="976537 noopener"&gt;VOSS-33244),&amp;nbsp;&lt;/A&gt;in case you don't want to take 9.3.0&lt;/P&gt;</description>
      <pubDate>Wed, 15 Oct 2025 11:06:20 GMT</pubDate>
      <guid>https://community.extremenetworks.com/t5/extremeswitching-vsp-fabric/high-cpu-on-subnet-scan/m-p/120519#M3103</guid>
      <dc:creator>Ludovico_Steven</dc:creator>
      <dc:date>2025-10-15T11:06:20Z</dc:date>
    </item>
  </channel>
</rss>

