Design & Build

Data Center Infrastructure-Validation Test: Brocade MLXe Multicast Compatibility, and Scalability with Informatica Messaging

by ‎01-16-2013 12:34 PM - edited ‎07-22-2014 09:08 AM (1,877 Views)

Synopsis: Validation test cases and results for the Brocade MLXe Router with the Informatica Messaging application in Foreign Exchange environments.

 

Contents

Preface

This document provides a validation test of the Brocade MLXe with the Informatica Messaging application. It demonstrates the multicast compatibility and scalability of the MLXe in a typical Foreign Exchange network solution.

 

Overview

A Foreign Exchange offers corporate and institutional buyers and sellers in the Foreign Exchange (FX), and money markets reliable, low-cost, secure electronic access to the $1+ trillion a day global FX market. This validation test demonstrates how the Brocade’s MLXe satisfies a Foreign Exchange’s L3 multicast compatibility and scale requirements when using the Informatica Messaging application. This goals for this validation test are:

  • Demonstrate MLXe L3 multicast compatibility, and scalability with the Informatica Messaging application
  • Demonstrate MLXe L3 multicast fast convergence time with the Informatica Messaging application
  • Demonstrate MLXe L3 multicast Rendezvous Point (RP) redundancy/failover using PIM Anycast RP

 

Purpose of This Document

To provide MLXe configuration, validation test cases and results with the Informatica Messaging application.

 

Audience

The validation test will be valuable to networking architects and designers who are responsible for configuring foreign exchange networks.

 

Conclusions

Refer to the validation test topology in Figure 1 for the location of equipment.

 

Test case

Result

Comments

MLXe Multicast Test Using Informatica Messaging Application  

  • In application stages 1-3, ips01/02.siteA and ips01/02.siteB publishes market data, book01.siteA and book01.siteB subscribed correctly.

Tested with customer’s Informatica Messaging application servers only.

MLXe Multicast Failover Test Using Informatica Messaging Application

  • Triggering failure by bringing down AR01.NY5 eth 1/5 connected to BSW01, 348 packets’ loss out of total 177541 packets was observed, application traffic flow resumed within 30 seconds on secondary NIC at “site A”, and application did not timeout during re-convergence.
  • Triggering recovery from the failure by bringing up AR01.NY5 eth 1/5 connected to BSW01, primary NIC at “site A”d, and 678 packets’ loss out of total 177688 packets.
  • Triggering failure by bringing down AR01.NY5 using reload CLI, application servers lost TCP connections, and resumed after 5 to 10 seconds interruption between each application. When AR01.NY5 returned back to service, no TCP session drop was observed.
  • Triggering failure by bringing down AR02.NY5 eth 1/5 connected to BSW02, no impact to application traffic was observed.
  • Triggering recovery from the failure by bringing up AR02.NY5 interface eth 1/5 connected to BSW02, no impact to application traffic was observed.
  • Triggering failure by bringing down AR02.NY5 using reload CLI, no impact to application traffic was observed. When AR01.NY5 returned back to service, no impact to application traffic was observed.
  • Triggering failure by bringing down CR01.NY5 eth 2/4, 181 packets’ loss out of total
  • Triggering recovery from the failure by bringing up CR01.NY5 eth 2/4, no impact to the application traffic was observed.
  • Triggering failure by shutting CR01.NY5 loopback 2 having PIM anycast RP’ address, 6 packets loss out of the total 8287 packets, and no impact to the application was observed.
  • Triggering recovery from the failure by unshutting CR01.NY5 loopback 2 having PIM anycast RP’ address, 8 packets’ loss out of total 178807 packets, and no impact to the applications were observed. 
  • Triggering by shutting CR01.NY4 loopback 2 having PIM anycast RP’ address, no packet loss, and no impact to application were observed. 
  • Triggering recovery from the failure by bringing up CR01.NY4 loopback 2 having PIM anycast RP’ address, no packet loss, and no impact to application were observed. 
  • Triggering failure by shutting the primary NIC of the server 1 in the “site A”, no packet loss, and no impact to application was observed. 

Tested with customer’s Informatica Messaging application servers only.

For failure scenarios, application packet loss, if application traffic flow resumes within 30 seconds, and if application does not timeout during re-convergence were measured.

MLXe Multicast Test Using Informatica Messaging Application under Network Stress

  • No impact to the application under this stress network environment of 6000 PIM mcache and IGMP snooping of 4000 groups in MLX.

Tested with customer’s Informatica Messaging application servers and STC.

To create network stress, for PIM SM, IGMP reports of 3000 groups were simulated on STC, creating 6000 PIM mcache entries in MLX. 

For IGMP snooping, IGMP report of 4000 groups more than max supported 2048 in MLX were simulated on STC.

MLXe Multicast Test Using Informatica Messaging Application under Extreme Network Stress

  • No obvious packet loss was observed from application’s perspective.                                    
  • Approximately 40% packets loss on STC and CPU hike of 24% on the management card of AR01.CH1 under extremely stressed network, but it is expected behavior since there are more mache entries than max supported in MLX.

Tested with customer’s Informatica Messaging application servers and STC.

To create extreme network stress, for PIM SM, IGMP reports of 6,000 groups were simulated on STC, creating 12000 PIM mcache entries more than max supported 8000 in MLX. 

For IGMP snooping, IGMP report of 4000 groups more than max supported 2048 in MLX were simulated on STC.

 

Document History

Date                    Version        Description

2013-01-17      1.0            Initial Version

2013-01-23      1.1            Updated graphics

 

Related Documents

The following documents are valuable resources for the designer. In addition, any Brocade release notes that have been published for NOS, FOS and the MLX NetIron should be reviewed.

 

References

 

Key Contributors

The content in this guide was provided by the following key contributors.

Test Architect:Chris Yoon

Test Engineer: Chris Yoon

 

About Brocade

Brocade® (NASDAQ: BRCD) networking solutions help the world’s leading organizations transition smoothly to a world where applications and information reside anywhere. This vision is designed to deliver key business benefits such as unmatched simplicity, non-stop networking, application optimization, and investment protection.

 

Innovative Ethernet and storage networking solutions for data center, campus, and service provider networks help reduce complexity and cost while enabling virtualization and cloud computing to increase business agility.

To help ensure a complete solution, Brocade partners with world-class IT companies and provides comprehensive education, support, and professional services offerings. (www.brocade.com)

 

About Informatica

Informatica Corporation is the world’s number one independent provider of data integration software. Thousands of enterprises worldwide depend on Informatica data integration, data quality, and big data solutions to access, integrate, and trust their information assets residing on premise and in the Cloud.

 

Test Plan

The validation test plan includes four test cases:

  1. Test Case #1:  Validate Configuration is Operating Correctly
  2. Test Case #2:  MLXe Multicast Failover Test Using Informatica Messaging Application
  3. Test Case #3:  MLXe Multicast Test using Informatica Messaging Application under Network Stress
  4. Test Case #4:  MLXe Multicast Test using Informatica Messaging Application under Extreme Network Stress

 

DUT Specifications

The table below lists the devices under test (DUT).

 

Identifier

Release

Configuration Options

Notes

BR-MLXE-4-AC

NI 5.3b.00

 

Four slot chassis, AC power

BR-MLXE-8-AC

NI 5.3b.00

 

Eight slot chassis, AC power

NI-MLX-4-AC

NI 5.3b.00

 

4-slot Brocade MLX-4 AC system

NI-MLX-8-AC

NI 5.3b.00

 

8-slot Brocade MLX-8 AC system

NI-MLX-MR

NI 5.3b.00

 

Brocade MLX system management module, 1 GB SDRAM, dual PCMCIA slots, EIA/TIA-232, and 10/100/1000 Ethernet

ports for out-of-band management

NI-X-16-8-HSF

NI 5.3b.00

 

Brocade MLX 8/16-slot system high-speed switch fabric module

NI-X-4-HSF

NI 5.3b.00

 

Brocade MLX 4-slot system high-speed switch fabric module

NI-MLX-10Gx8-M

NI 5.3b.00

 

Brocade MLX Series 8-port 10 GbE (M) module with IPv4/IPv6/MPLS hardware support. Support for up to 512,000 IPv4 routes. Requires SFP+ optics and high-speed switch fabric modules

NI-MLX-1Gx20-GC

NI 5.3b.00

 

Brocade MLX Series 20-port 10/100/1000 copper module with IPv4/IPv6/MPLS hardware support

NI-MLX-10Gx4

NI 5.3b.00

 

Brocade MLX Series 4-port 10 GbE module with IPv4/IPv6/MPLS hardware support

 

Test Equipment

Traffic was generated using a Spirent Test Center model SPT9000A with software release 3.90.

 

Test Configuration

As depicted in the Figure 1, the validation test topology consists of nine MLXe or MLX routers with the MR system management module used as layer-3 (L3) multicast PIM routers, two FCX switches as layer-2 (L2) multicast switches, and one IBM blade server chassis hosting the Informatica Messaging application. The IBM blade server chassis provides unicast client and server hosts, and multicast source and receiver hosts. The Spirent Test Center (STC) scales multicast traffic up to 4 Gbps to the multicast source and receiver.

 

Topology.jpg

Figure 1.  Validation Testing Topology (click to enlarge)

 

In the test topology, OSPF is enabled as IGP on all the MLXe/MLX routers, and OSPF area 0 is used. PIM-SM and static PIM-SM RP is enabled on all the MLXe & MLX routers. PIM Designated Router (DR) priority is configured for VE interfaces of the LH (Last Hop) router, AR01.CH1, and FH (First Hop) router, AR01.NY5, to be the DR for the other LH router, AR02.CH1, and FH router, AR02.NY5. Below is the OSPF and PIM-SM configuration for AR01.CH1.

--------------

telnet@AR01.CH1#

vlan 201

tagged ethe 1/1 ethe 2/3

router-interface ve 201

!

vlan 202

tagged ethe 1/2 ethe 2/3

router-interface ve 202

!

vlan 203

tagged ethe 1/3 ethe 2/3

router-interface ve 203

!

vlan 312

untagged ethe 1/6

tagged ethe 2/3 ethe 2/5

router-interface ve 312

!

!

router ospf

area 0

redistribute connected

log adjacency

!

router pim

rp-address 1.1.1.1 mcast-group-range

!

!

interface loopback 1

ip address 11.96.99.251/32

ip pim-sparse

!

!

interface ethernet 2/1

enable

ip ospf area 0

ip ospf network point-to-point

ip address 11.96.0.2/30

ip pim-sparse

!

interface ve 201

ip address 10.96.201.2/24

ip pim-sparse

ip pim dr-priority 2

ip vrrp-extended vrid 201

  backup

  ip-address 10.96.201.1                                         

  advertise backup

  activate

!

interface ve 202

ip address 10.96.202.2/24

ip pim-sparse

ip pim dr-priority 2

ip vrrp-extended vrid 202

  backup

  ip-address 10.96.202.1

  advertise backup

  activate

!

interface ve 203

ip address 10.96.203.2/24

ip pim-sparse

ip pim dr-priority 2

ip vrrp-extended vrid 203

  backup

  ip-address 10.96.203.1

  advertise backup

  activate

!

interface ve 312

ip address 10.96.112.2/24

ip pim-sparse

ip pim dr-priority 2

ip vrrp-extended vrid 31

  backup

  ip-address 10.96.112.1

  advertise backup

  activate

!

!

ip access-list standard mcast-group-range

permit 239.224.0.0 0.0.255.255

permit 239.225.0.0 0.0.255.255

permit 239.226.0.0 0.0.255.255

permit 224.23.0.0 0.0.255.255

permit 224.9.0.0 0.0.255.255

permit 239.227.0.0 0.0.255.255

permit 239.228.0.0 0.0.255.255

permit 239.229.0.0 0.0.255.255

!                                         

-------------

For PIM-SM RP, two anycast RPs, CR01.NY5 and CR01.NY4, are enabled as shown in Figure 1 to get fast convergence when a PIM-SM RP router fails. To allow receivers and sources to rendezvous to the closest RP, multicast packets from a source need to get to all anycast RPs by the PIM register mechanism as, shown in the Figure 2, to find the joined receivers:

 

image003.png

Figure 2.  PIM register mechanism with PIM anycast RP (click to enlarge)

 

Below is the PIM-SM anycast RP configuration of CR01.NY5 and CR01.NY4:

-------------

telnet@CR01.NY5#                  

router pim

rp-address 1.1.1.1 mcast-group-range

anycast-rp 1.1.1.1 10

!

!

interface loopback 1

ip ospf area 0

ip address 11.75.99.254/32

ip pim-sparse

!

interface loopback 2

ip ospf area 0

ip address 1.1.1.1/32

ip pim-sparse

!

!

access-list 10 permit host 11.75.99.254

access-list 10 permit host 11.46.99.254

!

ip access-list standard mcast-group-range

permit 239.224.0.0 0.0.255.255

permit 239.225.0.0 0.0.255.255

permit 239.226.0.0 0.0.255.255

permit 224.23.0.0 0.0.255.255

permit 224.9.0.0 0.0.255.255

permit 239.227.0.0 0.0.255.255                                  

permit 239.228.0.0 0.0.255.255

permit 239.229.0.0 0.0.255.255

telnet@CR01.NY4#

router pim

rp-address 1.1.1.1 mcast-group-range

anycast-rp 1.1.1.1 10

!

!

interface loopback 1

ip ospf area 0

ip address 11.46.99.254/32

ip pim-sparse

!

interface loopback 2

ip ospf area 0

ip address 1.1.1.1/32

ip pim-sparse

!

!

access-list 10 permit host 11.75.99.254

access-list 10 permit host 11.46.99.254

!

ip access-list standard mcast-group-range

permit 239.224.0.0 0.0.255.255

permit 239.225.0.0 0.0.255.255

permit 239.226.0.0 0.0.255.255

permit 224.23.0.0 0.0.255.255

permit 224.9.0.0 0.0.255.255

permit 239.227.0.0 0.0.255.255

permit 239.228.0.0 0.0.255.255

permit 239.229.0.0 0.0.255.255

------------

See the following reference for best practice information about configuring multicast PIM with Anycast RP.

 

References

 

As shown in Figure 3, the IBM blade servers 1-4 are connected to AR01.NY5 and AR02.NY5 via the layer 2 switches, BSW01 and BSW02, in the IBM blade server chassis. IBM blade servers 5-7 are connected to AR01.CH1 and AR02.CH2 in a similar manner as shown in the Figure 3.

 

TopologyWithServers.jpg

Figure3.  IBM blade server chassis’ connections to AR routers (click to enlarge)

 

Each server has two Network Interface Cards (NICs), and NIC Teaming is enabled on each server to allow two NICs to be placed into a team for traffic failover to prevent connectivity loss in the event of network component failure. In the NIC teaming configuration, the NIC connected to BSW01 is primary and the NIC connected to BSW02 is secondary. The secondary NIC is disabled when the primary NIC and 10GbE uplink of BSW01 is up. As soon as the primary NIC or the10GbE uplink of BSW goes down, the secondary NIC is automatically enabled. In the steady state, the IGMP report and other traffic from servers to the MLX routers, and vice versa, are forwarded through the primary NIC and switch BSW01, but not through switch BSW02.

Multicast sources and receivers reside in both “site A”, and “site B” so multicast traffic between “site A” and “site B” servers is bi-directional.

 

TopologyWithMulticastTraffic.jpg

Figure 4.  STC IGMP report and multicast traffic path (click to Enlarge)

 

As shown in the Figure 4, STC ports 4/1-3 are the multicast receivers that send IGMP reports for up to 6000 groups to layer-3 routers AR01.CH1 and AR02.CH1 via the FCX layer-2 switch. STC ports 4/5-7, are the multicast sources that send multicast traffic for up to 6000 groups to layer-3 routers AR01.NY5 and AR02.NY5, L3 via the FCX layer-2 switch. The multicast traffic path is

STC 4/5-7 --> FCX_125621 --> AR01.NY5 --> CR01.NY5 --> AR01.CH1 --> FCX624_122062 --> STC 4/1-3

 

When the layer-2 switch, FCX624_122062, receives the multicast traffic from AR01.CH1, DR, by default the FCX replicates the multicast traffic to all the ports other than the port it was received on. This behavior causes eth 2/1-2/3 of the non-DR router, AR02.CH1, to receive unwanted multicast traffic. Since the traffic does not hit the PIM mcache entries of AR02.CH1, the traffic goes to the CPU of line card 2, and is dropped. This causes the line card to experience high CPU utilization as shown below. IGMP and PIM-SM snooping are enabled for the FCX to prevent replicating the unwanted multicast traffic to eth 2/1-2/3 of AR02.CH1.

Below is the output of CPU utilization, interface traffic statistics, and PIM mcache in AR02.CH1 prior to IGMP and PIM-SM snooping being enabled in the FCX.

 

----------

telnet@AR02.CH1#sh cpu-utilization lp

SLOT  #:          LP CPU UTILIZATION in  %:

               in 1 second:  in 5 seconds: in 60 seconds: in 300 seconds:

     1:        1             1              1               1 

     2:        94            94             94              64

Eth 2/1 - 2/3 of AR02.CH1 receives unwanted multicast traffic, and hence,(S,G) entries without outgoing interface are created in the PIM mcache table:

telnet@AR02.CH1#sh statistics | in PO|Util

PORT 2/1 Counters:

    InUtilization             84.39%      OutUtilization               0.0%

PORT 2/2 Counters:

    InUtilization             84.39%      OutUtilization               0.0%

PORT 2/3 Counters:

InUtilization             85.29%      OutUtilization               0.0%

telnet@AR02.CH1# sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 3000

1     (11.75.112.3, 239.224.1.1) in e1/1 (e1/1), Uptime 00:00:40, Rate 0 (SM)

      upstream neighbor 11.96.0.5

      Flags (0x000200c1) SM HW FAST

      fast ports:

      AgeSltMsk: 00000001,  FID: 0x8000 (D), MVID: NotReq, AvgRate: 0, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0

2     (11.75.112.3, 239.224.1.2) in e1/1 (e1/1), Uptime 00:00:40, Rate 0 (SM)

      upstream neighbor 11.96.0.5

      Flags (0x000200c1) SM HW FAST

      fast ports:

      AgeSltMsk: 00000001,  FID: 0x8000 (D), MVID: NotReq, AvgRate: 0, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0     

----------

Below is the IGMP and PIM-SM snooping configuration in the FCX:

----------

telnet@ FCX648_122062

vlan 112 by port

tagged ethe 1/1/1 ethe 1/1/5 ethe 1/1/9

multicast passive

multicast pimsm-snooping

!

vlan 113 by port

tagged ethe 1/1/2 ethe 1/1/6 ethe 1/1/10

multicast passive

multicast pimsm-snooping

!

vlan 114 by port

tagged ethe 1/1/3 ethe 1/1/7 ethe 1/1/11

multicast passive

multicast pimsm-snooping

!

!

system-max igmp-max-group-addr 8192

system-max igmp-snoop-mcache 8192

----------

 

Test Cases

Test Case #1: Validate Configuration is Operating Correctly

DUT

This is the test configuration diagram.

 

TopologyWithServers.jpg

Figure 5.  Test Case #1 Configuration (click to enlarge)

 

Purpose

To validate ips01.siteA is publishing market data, book01.siteA and book01.siteB are subscribing. Site A is formed of servers 5-7, and site B is formed of servers 1-4 as shown in the Test #1 Configuration.

 

Test Procedure

Step 1: Configure site A

For “site A”, configure eth 2/5 & eth 2/3 of AR01.CH1 and eth 1/5 & eth 1/3 of AR02.CH1 to be VLAN 312 as depicted in the Test#1 Configuration.

 

Step 2: Configure Site B

Configure for “site B”,     eth 1/5 & eth 1/3 of AR01.NY5 and eth 1/5 & eth 1/3 of AR02.NY5 in the same way.  Following is VLAN & VE configuration in AR01.CH1, AR02.CH1, AR01.NY5, and AR02.NY5:

---------

 

telnet@AR01.CH1#

vlan 312

tagged ethe 2/3 ethe 2/5

router-interface ve 312

no spanning-tree

!

ip multicast-routing

ip multicast-routing fast-convergence

!

interface ve 312

ip address 10.96.112.2/24

ip pim-sparse

ip pim dr-priority 2

ip vrrp-extended vrid 31

  backup

  ip-address 10.96.112.1

  advertise backup

activate

telnet@AR02.CH1#

vlan 312

tagged ethe 1/3 ethe 1/5

router-interface ve 312

no spanning-tree

!

ip multicast-routing

ip multicast-routing fast-convergence

!

interface ve 312

ip address 10.96.112.3/24

ip pim-sparse

ip vrrp-extended vrid 31

  backup priority 95

ip-address 10.96.112.1

advertise backup

activate

telnet@AR01.NY5#

!

vlan 112

tagged ethe 1/3 ethe 1/5                                        

router-interface ve 112                                         

no spanning-tree

!

vlan 113                                                         

tagged ethe 1/3 ethe 1/5                                        

router-interface ve 113                                         

no spanning-tree

!                                                                

  ip multicast-routing

  ip multicast-routing fast-convergence

!

interface ve 112

ip address 10.75.112.2/24                                       

ip pim-sparse

ip pim dr-priority 2                                            

ip vrrp-extended vrid 112                                       

backup

ip-address 10.75.112.1                                         

advertise backup

activate

!

interface ve 113

ip address 10.75.113.2/24                                       

ip pim-sparse

ip pim dr-priority 2                                            

ip vrrp-extended vrid 113                                       

backup

ip-address 10.75.113.1                                         

advertise backup

activate

telnet@AR02.NY5#

vlan 112

tagged ethe 1/3 ethe 1/5                                        

router-interface ve 112                                         

no spanning-tree                                                

!

vlan 113

tagged ethe 1/3 ethe 1/5                                        

router-interface ve 113                                         

no spanning-tree

!

ip multicast-routing                                             

ip multicast-routing fast-convergence                            

!

interface ve 112

ip address 10.75.112.3/24                                       

ip pim-sparse

ip vrrp-extended vrid 112                                       

  backup priority 95                                             

  ip-address 10.75.112.1                                         

  advertise backup

  activate

!

interface ve 113                                                 

ip address 10.75.113.3/24                                       

ip pim-sparse

ip vrrp-extended vrid 113                                       

  backup priority 95                                             

  ip-address 10.75.113.1                                         

  advertise backup

  activate                    

----------

 

Step 3: Start Informatica Messaging application

Start the Informatica Messaging application on the servers in “site A” and “site B” for ips01.siteA to publish market data, and for book01.siteA and book01.siteB to subscribe the market data.

 

Result

1. In the application stages 1-3, ips01/02.siteA and ips01/02.siteB are publishing market data, book01.siteA and book01.siteB are subscribing correctly. Following is the output of the show commands about IGMP group, and PIM mcache etc.

 

----------

telnet@AR01.CH1#sh ip igmp group

Total 65 groups

--------------------------------------------------------------

Idx   Group Address    Port Intf   GrpCmpV Mode    Timer Srcs

----+----------------+------+------+-------+-------+-----+----

   1 224.9.10.23      e2/5 v312      Ver2 exclude   176 0

   2 224.10.10.21     e2/5 v312      Ver2 exclude   179 0

   3 224.23.10.2      e2/5 v312      Ver2 exclude   178 0

   4 224.23.10.3      e2/5 v312      Ver2 exclude   176 0

   5 224.23.10.4      e2/5 v312      Ver2 exclude   182    0

telnet@AR01.CH1#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 231

1     (*, 224.9.10.23) RP 1.1.1.1, in e2/1 (NIL), Uptime 01:07:24 (SM)

      upstream neighbor 11.96.0.1

      Flags (0x002604a0) SM RPT LRCV TAG

      slow ports: ethe 2/5

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

e2/5(VL312), 01:06:10/0, Flags: MJ

2     (10.75.112.20, 224.9.10.23) in e2/1 (e2/1), Uptime 01:09:01, Rate 32 (SM)

      upstream neighbor 11.96.0.1

      Flags (0x700680e1) SM SPT LRCV HW FAST TAG

      fast ports: ethe 2/5                                       

      AgeSltMsk: 00000002,  FID: 0x8343, MVID:    28 , AvgRate: 32, profile: none

      Forwarding_oif: 1, Immediate_oif: 0, Blocked_oif: 0        

      L3 (HW) 1:

          e2/5(VL312), 01:06:10/0, Flags: MJ

5     (10.96.112.20, 224.9.10.23) in v312 (tag e2/5), Uptime 01:09:03, Rate 51 (SM)

      Source is directly connected. RP 1.1.1.1                   

      Flags (0xb046ace1) SM SPT L2REG REGSUPP LSRC LRCV HW FAST TAG

      fast ports: ethe 2/1                                       

      AgeSltMsk: 00000002,  FID: 0x835a, MVID:    23 , AvgRate: 51, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

          TR(e2/1,e2/1)(VL1), 01:08:43/166, Flags: IM            

      Blocked OIF 1:                                             

          e2/5(VL312), 01:06:11/0, Flags: MJ BI  

telnet@AR01.CH1#sh cpu-utilization  

... Usage average for all tasks in the last 1 seconds  ...

==========================================================

Name                    us/sec          %

idle                    973828          99

ip_rx                   10703           1

telnet@AR01.CH1#sh cpu-utilization lp

SLOT  #:          LP CPU UTILIZATION in  %:

               in 1 second:  in 5 seconds: in 60 seconds: in 300 seconds:

     1:        1             1              1               1 

     2: 1             1              1               1 

----------

Since server 1-4’ NICs connected to BSW-2 are disabled in the steady state, AR02.CH1 does not receive any IGMP report, so (S,G) entries does not have any OIL interface as shown below:

----------

telnet@AR02.CH1#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 102

1     (10.96.112.20, 224.9.10.23) in v312 (tag e1/3), Uptime 01:11:31 (SM)

      upstream neighbor is L2 10.96.112.2. RP 1.1.1.1

      Flags (0x004280c0) SM SPT LSRC HW

      slow ports:

      AgeSltMsk: 00000000,  FID: 0x8000 (D), MVID: NotReq, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0

2     (10.96.112.21, 224.9.10.23) in v312 (tag e1/3), Uptime 01:11:31 (SM)

      upstream neighbor is L2 10.96.112.2. RP 1.1.1.1

      Flags (0x004280c0) SM SPT LSRC HW

      slow ports:

      AgeSltMsk: 00000000,  FID: 0x8000 (D), MVID: NotReq, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0

telnet@AR02.CH1_MLXe4_134671#sh ip igmp group

Total 0 groups

telnet@AR02.CH1#sh cpu

... Usage average for all tasks in the last 1 seconds  ...

==========================================================

Name                    us/sec          %

idle                    981470          100

telnet@AR02.CH1#sh cpu-utilization lp

SLOT  #:          LP CPU UTILIZATION in  %:

               in 1 second:  in 5 seconds: in 60 seconds: in 300 seconds:

     1:        1             1              1               1 

     2:        1             1              1               1 

telnet@CR01.CH1#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 228

1     (*, 224.9.10.23) RP 1.1.1.1, in e2/4 (NIL), Uptime 01:11:09 (SM)

      upstream neighbor 11.100.100.1

      Flags (0x00220480) SM RPT

      slow ports: ethe 2/1

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

TR(e2/1,e2/1)(VL1), 01:07:50/151, Flags: IM

2     (10.75.112.20, 224.9.10.23) in e2/4 (e2/4), Uptime 01:12:47, Rate 32 (SM)

      upstream neighbor 11.100.100.1

      Flags (0x300284c1) SM SPT HW FAST

      fast ports: ethe 2/2                                       

      AgeSltMsk: 00000002,  FID: 0x8003, MVID: NotReq, AvgRate: 32, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0        

      L3 (HW) 1:

TR(e2/1,e2/2)(VL1), 01:07:50/151, Flags: IM IH                           

5     (10.96.112.20, 224.9.10.23) in e2/1 (e2/1), Uptime 01:12:37, Rate 52 (SM)

      upstream neighbor 11.96.0.2

      Flags (0xb00284c1) SM SPT HW FAST                          

      fast ports: ethe 2/4                                       

      AgeSltMsk: 00000002,  FID: 0x8005, MVID: NotReq, AvgRate: 51, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

e2/4(VL1), 01:12:37/185, Flags: IM                     

      Blocked OIF 1:                                             

          TR(e2/1,e2/1)(VL1), 01:07:52/209, Flags: IH BR BI      

telnet@CR01.CH1#sh cpu

... Usage average for all tasks in the last 1 seconds  ...

==========================================================

Name                    us/sec          %

idle                    981356          100

telnet@CR01.CH1#sh cpu lp

SLOT  #:          LP CPU UTILIZATION in  %:

               in 1 second:  in 5 seconds: in 60 seconds: in 300 seconds:

     2:        1             1              1               1 

telnet@CR02.CH1#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 0

mcache not found

telnet@CR01.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 288

1     (*, 224.9.10.23) RP 1.1.1.1, in NIL (NIL), Uptime 01:13:44 (SM)

      No upstream neighbor because RP 1.1.1.1 is itself

      Flags (0x00220480) SM RPT

      slow ports: ethe 2/1 ethe 2/4

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, RegPkt: 0, profile: none

      Forwarding_oif: 2, Immediate_oif: 2, Blocked_oif: 0

      L3 (SW) 2:

          TR(e2/1,e2/1)(VL1), 01:13:44/166, Flags: IM

e2/4(VL1), 01:08:59/175, Flags: IM

2     (10.75.112.20, 224.9.10.23) in e2/1 (e2/1), Uptime 01:13:47, Rate 31 (SM)

      upstream neighbor 11.75.0.2

      Flags (0xb002c4c1) SM SPT HW FAST MSDPADV                  

      fast ports: ethe 2/4                                       

      AgeSltMsk: 00000002,  FID: 0x8006, MVID: NotReq, RegPkt: 0, AvgRate: 32, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

e2/4(VL1), 01:08:59/175, Flags: IM IH                  

      Blocked OIF 1:                                             

          TR(e2/1,e2/1)(VL1), 01:13:44/166, Flags: IH BR BI      

5     (10.96.112.20, 224.9.10.23) in e2/4 (e2/4), Uptime 01:13:49, Rate 51 (SM)

      upstream neighbor 11.100.100.2

      Flags (0xb002c4c1) SM SPT HW FAST MSDPADV                  

      fast ports: ethe 2/1                                       

      AgeSltMsk: 00000002,  FID: 0x8003, MVID: NotReq, RegPkt: 0, AvgRate: 51, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

          TR(e2/1,e2/1)(VL1), 01:13:49/161, Flags: IM IH         

      Blocked OIF 1:                                             

          e2/4(VL1), 01:09:05/170, Flags: IH BR BI   

telnet@CR02.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 0

mcache not found

----------

CR01.NY4 of a PIM anycast RP has received PIM register packets from CR01.NY5 that is another PIM anycast RP, but it does not have any (*,G) entries since it haven’t received any PIM join: 

----------

telnet@CR01.NY4#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 210

1     (10.75.112.20, 224.9.10.23) in e2/1 (e2/1), Uptime 01:14:10 (SM)

      upstream neighbor 11.100.100.9

      Flags (0x000240c0) SM HW MSDPADV

      slow ports:

      AgeSltMsk: 00000000,  FID: 0x8000 (D), MVID: NotReq, RegPkt: 0, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0

2     (10.75.112.21, 224.9.10.23) in e2/1 (e2/1), Uptime 01:15:03 (SM)

      upstream neighbor 11.100.100.9

      Flags (0x000240c0) SM HW MSDPADV

      slow ports:

      AgeSltMsk: 00000000,  FID: 0x8000 (D), MVID: NotReq, RegPkt: 0, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0

telnet@AR01.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 243

1     (*, 224.9.10.23) RP 1.1.1.1, in e1/1 (NIL), Uptime 01:15:19 (SM)

      upstream neighbor 11.75.0.1

      Flags (0x002604a0) SM RPT LRCV TAG

      slow ports: ethe 1/5

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          e1/5(VL112), 01:15:19/0, Flags: MJ

2     (10.75.112.20, 224.9.10.23) in v112 (tag e1/5), Uptime 01:15:54, Rate 31 (SM)

      Source is directly connected. RP 1.1.1.1

      Flags (0xb046ace1) SM SPT L2REG REGSUPP LSRC LRCV HW FAST TAG

      fast ports: ethe 1/1                                       

      AgeSltMsk: 00000001,  FID: 0x8008, MVID:     9 , AvgRate: 32, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

TR(e1/1,e1/1)(VL1), 01:11:19/207, Flags: IM            

      Blocked OIF 1:                                             

          e1/5(VL112), 01:15:19/0, Flags: MJ BI                  

3     (10.75.112.21, 224.9.10.23) in v112 (tag e1/5), Uptime 01:15:54, Rate 33 (SM)

      Source is directly connected. RP 1.1.1.1                   

      Flags (0xb046ace1) SM SPT L2REG REGSUPP LSRC LRCV HW FAST TAG

      fast ports: ethe 1/2                                       

      AgeSltMsk: 00000001,  FID: 0x800a, MVID:    10 , AvgRate: 35, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

          TR(e1/1,e1/2)(VL1), 01:11:19/207, Flags: IM            

      Blocked OIF 1:                                             

          e1/5(VL112), 01:15:19/0, Flags: MJ BI                  

telnet@AR01.NY5# sh ip igmp group

Total 67 groups

--------------------------------------------------------------

Idx   Group Address    Port Intf   GrpCmpV Mode    Timer Srcs

----+----------------+------+------+-------+-------+-----+----

   1 224.9.10.23      e1/5 v112      Ver2 exclude   222 0

   2 224.10.10.21     e1/5 v112      Ver2 exclude   219 0

   3 224.23.10.2      e1/5 v112      Ver2 exclude   219 0

telnet@AR02.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP,  PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 108

1 (10.75.112.20, 224.9.10.23) in v112 (tag e1/3), Uptime 01:16:10 (SM)

upstream neighbor is L2 10.75.112.2. RP 1.1.1.1

Flags (0x004280c0) SM SPT LSRC HW

slow ports:

AgeSltMsk: 00000000,  FID: 0x8000 (D), MVID: NotReq, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0

2 (10.75.112.21, 224.9.10.23) in v112 (tag e1/3), Uptime 01:16:10 (SM)

upstream neighbor is L2 10.75.112.2. RP 1.1.1.1

Flags (0x004280c0) SM SPT LSRC HW

slow ports:

AgeSltMsk: 00000000,  FID: 0x8000 (D), MVID: NotReq, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0        

----------

Multicast traffic load-balacing is working on LAG member links of eth 1/1 and eth 1/2:  

----------

telnet@AR01.NY5#show int ethernet 1/1

10GigabitEthernet1/1 is up, line protocol is up

  STP Root Guard is disabled, STP BPDU Guard is disabled

  Hardware is 10GigabitEthernet, address is 0024.389a.a100 (bia 0024.389a.a100)

  Configured speed 10Gbit, actual 10Gbit, configured duplex fdx, actual fdx

  Member of Control VLAN 4095, VLAN 1 (untagged), port is in untagged mode, port state is Forwarding

  STP configured to ON, Priority is level0, flow control enabled

  Priority force disabled, Drop precedence level 0, Drop precedence force disabled

  dhcp-snooping-trust configured to OFF

  mirror disabled, monitor disabled

  LACP BPDU ForwardingSmiley Very Happyisabled

  LLDP BPDU ForwardingSmiley Very Happyisabled

  Member of active trunk ports 1/1-1/2, primary port

  Member of configured trunk ports 1/1-1/2, primary port

  No port name

  Port is not enabled to receive all vlan packets for pbr

  Internet address is 11.75.0.2/30, MTU 1548 bytes, encapsulation ethernet

  Cluster L2 protocol forwarding enabled

  300 second input rate: 967789 bits/sec, 533 packets/sec, 0.01% utilization

  300 second output rate: 1323714 bits/sec, 761 packets/sec, 0.01% utilization

telnet@AR01.NY5#   sh int ethernet 1/2

10GigabitEthernet1/2 is up, line protocol is up

  STP Root Guard is disabled, STP BPDU Guard is disabled

  Hardware is 10GigabitEthernet, address is 0024.389a.a100 (bia 0024.389a.a101)

  Configured speed 10Gbit, actual 10Gbit, configured duplex fdx, actual fdx

  Member of Control VLAN 4095, VLAN 1 (untagged), port is in untagged mode, port state is Forwarding

  STP configured to ON, Priority is level0, flow control enabled

  Priority force disabled, Drop precedence level 0, Drop precedence force disabled

  dhcp-snooping-trust configured to OFF

  mirror disabled, monitor disabled

  LACP BPDU ForwardingSmiley Very Happyisabled

  LLDP BPDU ForwardingSmiley Very Happyisabled

  Member of active trunk ports 1/1-1/2, secondary port, primary port is 1/1

  Member of configured trunk ports 1/1-1/2, secondary port, primary port is 1/1

  No port name

  Port is not enabled to receive all vlan packets for pbr

  MTU 1548 bytes, encapsulation ethernet

  Cluster L2 protocol forwarding enabled

  300 second input rate: 1200276 bits/sec, 614 packets/sec, 0.01% utilization

  300 second output rate: 1133941 bits/sec, 734 packets/sec, 0.01% utilization 

----------

Test Case #2: MLXe Multicast Failover Test Using Informatica Messaging Application

 

DUT

This is the test configuration diagram.

 

TopologyWithServers.jpg

Figure 6.  Test Case #2 Configuration (click to enlarge)

 

Purpose

To measure multicast flow recovery time, traffic flow resumed within 30 seconds, and application did not timeout during re-convergence.

 

Test Procedure

Step 1:The same as step 1 in Test Case #1.

 

Step 2: The same as step 2 in Test Case #1.

 

Step 3: Measure packet loss for various failure conditions

For the following failure conditions, measure application’s packet loss, application traffic flow resumes within 30 seconds, and application does not timeout during re-convergence.

1. Failure of AR01.NY5, LH & FH router:

    • Bring down interface eth 1/5 connected to BSW01
    • Bring up interface eth 1/5 connected to BSW01
      • Router reload

2. Failure of AR02.NY5, LH & FH router:

    • Bring down interface eth 1/5 connected to BSW02
    • Bring up interface eth 1/5 connected to BSW02
      • Router reload

3. Failure of CR01.NY5, core router, in the primary path and PIM-SM anycast RP:

    • Bring down interface eth 2/4 connected to CR01.CH1
    • Bring up interface eth 2/4 connected to CR01.CH1
    • Shut for loopback 2 having anycast RP’ address
    • Unshut for loopback 2 having anycast RP’ address

4. Failure of CR01.NY4, core router, in the backup path and PIM-SM anycast RP:

  • Shut for loopback 2 having anycast RP’ address
  • Unshut for loopback 2 having anycast RP’ address

5. Failure of the primary NIC of the server 1 in the “site A”:

  • Shut the primry NIC of the server 1 in the “site A”
  • Unshut the primary NIC of the server 1 in the “site A”

Results

1. Triggering failure by bringing down AR01.NY5 eth 1/5 connected to BSW01, 348 packets’ loss out of total 177541 packets was observed, application traffic flow resumed within 30 seconds on secondary NIC at “site A” servers, and application did not timeout during re-convergence. Below is captured output after AR01.NY5 eth 1/5 was brought down:

 

------------

telnet@AR01.NY5#sh ip igmp group

Total 0 groups

telnet@AR02.NY5#sh ip pim mcache                                    

IP Multicast Mcache Table

Entry Flags    : SM - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 243

1     (*, 224.9.10.23) RP 1.1.1.1, in e1/1 (NIL), Uptime 00:03:42 (SM)

      upstream neighbor 11.75.0.5

      Flags (0x002604a0) SM RPT LRCV TAG

      slow ports: ethe 1/5

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          e1/5(VL112), 00:03:42/0, Flags: MJ

2     (10.75.112.20, 224.9.10.23) in v112 (tag e1/5), Uptime 01:30:15, Rate 32 (SM)

      Source is directly connected. RP 1.1.1.1

      Flags (0xb046acc1) SM SPT L2REG REGSUPP LSRC LRCV HW FAST

      fast ports: ethe 1/3                                       

      AgeSltMsk: 00000001,  FID: 0x800f, MVID:    12 , AvgRate: 27, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L2 (HW) 1:

          e1/3, 00:03:41/203, Flags: IM                         

      Blocked OIF 1:                                             

          e1/5(VL112), 00:03:42/0, Flags: MJ BI

----------

2.  Triggering recovery from the failure by bringing up AR01.NY5 eth 1/5 connected to BSW01, primary NIC at “site A” servers resumed to service within 30 seconds, and 678 packets’ loss out of total 177688 packets was observed. Below is captured output after AR01.NY5 eth 1/5 was brought up:

-----------

telnet@AR01.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 243

1     (*, 224.9.10.23) RP 1.1.1.1, in e1/1 (NIL), Uptime 00:05:03 (SM)

      upstream neighbor 11.75.0.1

      Flags (0x002604a0) SM RPT LRCV TAG

      slow ports: ethe 1/5

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          e1/5(VL112), 00:05:03/0, Flags: MJ

2     (10.75.112.20, 224.9.10.23) in v112 (tag e1/5), Uptime 01:49:15, Rate 32 (SM)

      Source is directly connected. RP 1.1.1.1

      Flags (0xb046ace1) SM SPT L2REG REGSUPP LSRC LRCV HW FAST TAG

      fast ports: ethe 1/1                                       

      AgeSltMsk: 00000001,  FID: 0x8008, MVID:     9 , AvgRate: 32, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

          TR(e1/1,e1/1)(VL1), 01:44:40/197, Flags: IM            

      Blocked OIF 1:                                             

          e1/5(VL112), 00:05:03/0, Flags: MJ BI                  

3     (10.75.112.21, 224.9.10.23) in v112 (tag e1/5), Uptime 01:49:15, Rate 34 (SM)

      Source is directly connected. RP 1.1.1.1                   

      Flags (0xb046ace1) SM SPT L2REG REGSUPP LSRC LRCV HW FAST TAG

      fast ports: ethe 1/2                                       

      AgeSltMsk: 00000001,  FID: 0x800a, MVID:    10 , AvgRate: 35, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

          TR(e1/1,e1/2)(VL1), 01:44:40/197, Flags: IM            

      Blocked OIF 1:                                             

          e1/5(VL112), 00:05:03/0, Flags: MJ BI                  

telnet@AR02.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 177

1     (10.75.112.20, 224.9.10.23) in v112 (tag e1/3), Uptime 01:49:03, Rate 0 (SM)

      upstream neighbor is L2 10.75.112.2. RP 1.1.1.1

      Flags (0x304280c1) SM SPT LSRC HW FAST

      fast ports:

      AgeSltMsk: 00000000,  FID: 0x8000 (D), MVID: NotReq, AvgRate: 0, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0

2     (10.75.112.21, 224.9.10.23) in v112 (tag e1/3), Uptime 01:49:03, Rate 0 (SM)

      upstream neighbor is L2 10.75.112.2. RP 1.1.1.1

      Flags (0x304280c1) SM SPT LSRC HW FAST

      fast ports:

      AgeSltMsk: 00000000,  FID: 0x8000 (D), MVID: NotReq, AvgRate: 0, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 0                                            

----------

3. Triggering failure by bringing down AR01.NY5 using reload CLI, application servers lost TCP connections, and resumed after 5 to 10 seconds interruption between each application. When AR01.NY5 returned back to service, no TCP session drop was observed.

4. Triggering failure by bringing down AR02.NY5 eth 1/5 connected to BSW02, no impact to application traffic was observed.

5. Triggering recovery from the failure by bringing up AR02.NY5 interface eth 1/5 connected to BSW02, no impact to application traffic was observed.

6. Triggering failure by bringing down AR02.NY5 using reload CLI, no impact to application traffic was observed. When AR01.NY5 returned back to service, no impact to application traffic was observed.

7. Triggering failure by bringing down CR01.NY5 eth 2/4, 181 packets’ loss out of total 292351 packets was observed, traffic flow resumed within 30 seconds, and application did not timeout during re-convergence.

-------------

telnet@CR01.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 287

1     (*, 224.9.10.23) RP 1.1.1.1, in NIL (NIL), Uptime 02:23:06 (SM)

      No upstream neighbor because RP 1.1.1.1 is itself

      Flags (0x00220480) SM RPT

      slow ports: ethe 2/1

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, RegPkt: 0, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          TR(e2/1,e2/1)(VL1), 00:28:24/206, Flags: IM

2     (10.75.112.20, 224.9.10.23) in e2/1 (e2/1), Uptime 02:23:09, Rate 31 (SM)

      upstream neighbor 11.75.0.2

      Flags (0xb002c0c1) SM SPT HW FAST MSDPADV

      fast ports:

      AgeSltMsk: 00000002,  FID: 0x8000 (D), MVID: NotReq, RegPkt: 0, AvgRate: 32, profile: none

      Forwarding_oif: 0, Immediate_oif: 0, Blocked_oif: 1        

      Blocked OIF 1:                                             

          TR(e2/1,e2/1)(VL1), 00:28:24/206, Flags: IH BR BI      

telnet@AR01.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 243

1     (*, 224.9.10.23) RP 1.1.1.1, in e1/1 (e1/1), Uptime 00:30:44 (SM)

      upstream neighbor 11.75.0.1

      Flags (0x002604a0) SM RPT LRCV TAG

      slow ports: ethe 1/5

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          e1/5(VL112), 00:30:44/0, Flags: MJ

2     (10.75.112.20, 224.9.10.23) in v112 (tag e1/5), Uptime 00:31:20, Rate 33 (SM)

      Source is directly connected. RP 1.1.1.1

      Flags (0xb046ace1) SM SPT L2REG REGSUPP LSRC LRCV HW FAST TAG

      fast ports: ethe 1/1 ethe 1/3                              

AgeSltMsk: 00000001,  FID: 0x800c, MVID:     6 , AvgRate: 32, profile: none

      Forwarding_oif: 2, Immediate_oif: 2, Blocked_oif: 1        

      L3 (HW) 1:

          TR(e1/1,e1/1)(VL1), 00:31:07/15, Flags: IM             

      L2 (HW) 1:

          e1/3,  00:02:28/158, Flags: IM                         

      Blocked OIF 1:                                             

          e1/5(VL112), 00:30:44/0, Flags: MJ BI

-------------

8. Triggering recovery from the failure by bringing up CR01.NY5 eth 2/4, no impact to the application traffic was observed.

9. Triggering failure by shutting CR01.NY5 loopback 2 having PIM anycast RP’ address,  6 packets loss out of the total 8287 packets, and no impact to application was observed.

----------

telnet@CR01.CH1#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 228

1     (*, 224.9.10.23) RP 1.1.1.1, in e2/3 (e2/3), Uptime 02:34:46 (SM)

      upstream neighbor 11.96.0.10

      Flags (0x00220480) SM RPT

      slow ports: ethe 2/1

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          TR(e2/1,e2/1)(VL1), 02:31:28/174, Flags: IM

2     (10.75.112.20, 224.9.10.23) in e2/4 (e2/4), Uptime 02:36:25, Rate 31 (SM)

      upstream neighbor 11.100.100.1

      Flags (0x300284c1) SM SPT HW FAST

      fast ports: ethe 2/1                                       

      AgeSltMsk: 00000002,  FID: 0x8006, MVID: NotReq, AvgRate: 32, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0        

      L3 (HW) 1:

          TR(e2/1,e2/1)(VL1), 02:31:28/174, Flags: IM IH         

telnet@CR01.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 243

1     (*, 224.9.10.23) RP 1.1.1.1, in e2/3 (e2/3), Uptime 02:40:58 (SM)

      upstream neighbor 11.75.0.10

      Flags (0x00220480) SM RPT

      slow ports: ethe 2/1

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          TR(e2/1,e2/1)(VL1), 00:46:16/154, Flags: IM

2     (10.75.112.20, 224.9.10.23) in e2/1 (e2/1), Uptime 02:41:01, Rate 33 (SM)

      upstream neighbor 11.75.0.2

      Flags (0xb002c4c1) SM SPT HW FAST MSDPADV

      fast ports: ethe 2/4                                       

      AgeSltMsk: 00000002,  FID: 0x8003, MVID: NotReq, AvgRate: 31, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

          e2/4(VL1), 00:09:05/163, Flags: IM                     

      Blocked OIF 1:                                             

          TR(e2/1,e2/1)(VL1), 00:46:16/154, Flags: IH BR BI   

----------

10. Triggering recovery from the failure by unshutting CR01.NY5 loopback 2 having PIM anycast RP’ address, 8 packets’ loss out of total 178807 packets, and no impact to the applications were observed.

----------

telnet@CR01.CH1#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 228

1     (*, 224.9.10.23) RP 1.1.1.1, in e2/4 (e2/4), Uptime 02:49:09 (SM)

      upstream neighbor 11.100.100.1

      Flags (0x00220480) SM RPT

      slow ports: ethe 2/1

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

      Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          TR(e2/1,e2/1)(VL1), 02:45:51/151, Flags: IM

telnet@CR01.NY5#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags    : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

                 RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

                 HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

                 REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

                 MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP, PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

                 MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

                 BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 288

1     (*, 224.9.10.23) RP 1.1.1.1, in NIL (NIL), Uptime 02:52:56 (SM)

      No upstream neighbor because RP 1.1.1.1 is itself

      Flags (0x00220482) SM RPT

      slow ports: ethe 2/1 ethe 2/4

      AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, RegPkt: 0, profile: none

      Forwarding_oif: 2, Immediate_oif: 2, Blocked_oif: 0

      L3 (SW) 2:

          TR(e2/1,e2/1)(VL1), 00:58:13/156, Flags: IM

          e2/4(VL1), 00:04:24/163, Flags: IM

----------

11. Triggering by shutting CR01.NY4 loopback 2 having PIM anycast RP’ address, no packet loss, and no impact to application were observed.

12. Triggering recovery from the failure by unshutting CR01.NY4 loopback 2 having PIM anycast RP’ address, no packet loss, and no impact to application were observed.

13. Triggering failure by shutting the primary NIC of the server 1 in the “site A”, no packet loss, and no impact to application was observed.

 

Test Case #3: MLXe Multicast Test using Informatica Messaging Application under Network Stress

DUT

This is the test configuration diagram.

 

TopologyWithMulticastTraffic.jpg

Figure 7.  Test Case #3 Configuration (click to enlarge)

 

Purpose

To measure application’s behavior under stress network environment, PIM mcache table, and IP multicast resource allocation table.

 

Test Procedure

Step 1: The same as step 1 in Test Case #1.

 

Step 2: The same as step 2 in Test Case #1.

 

Step 3: Simulate network stress condition

STC 4/1-3 simulates IGMP report of 3000 groups, and STC 4/5-7 generates 3 x 800Mbps multicast traffic streams of the 3000 groups that create 6000 PIM mcache entries of 3000 (*,G) and 3000 (S,G) entries each on AR01.CH1.

 

----------

telnet@AR01.CH1#sh ip igmp group

Total 3065 groups

--------------------------------------------------------------

Idx   Group Address    Port Intf   GrpCmpV Mode    Timer Srcs

----+----------------+------+------+-------+-------+-----+----

   1 224.9.10.23      e2/5 v312      Ver2 exclude   145 0

   2 224.10.10.21     e2/5 v312      Ver2 exclude   145 0

   3 224.23.10.2      e2/5 v312      Ver2 exclude   149 0

   4 224.23.10.3      e2/5 v312      Ver2 exclude   146 0

   5 224.23.10.4      e2/5 v312      Ver2 exclude   150 0

telnet@AR01.CH1#sh ip pim mcache

IP Multicast Mcache Table

Entry Flags : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP,  PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 6231

1     (*, 224.9.10.23) RP 1.1.1.1, in e2/1 (e2/1), Uptime 19:57:59 (SM)

upstream neighbor 11.96.0.1

Flags (0x002604a0) SM RPT LRCV TAG

slow ports: ethe 2/5

AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

          e2/5(VL312), 19:56:45/0, Flags: MJ

2 (10.75.112.20, 224.9.10.23) in e2/1 (e2/1), Uptime 19:59:36, Rate 31 (SM)

upstream neighbor 11.96.0.1

Flags (0x700680e1) SM SPT LRCV HW FAST TAG

fast ports: ethe 2/5                                       

AgeSltMsk: 00000002,  FID: 0x8343, MVID:    28 , AvgRate: 31, profile: none

Forwarding_oif: 1, Immediate_oif: 0, Blocked_oif: 0        

      L3 (HW) 1:                                                 

          e2/5(VL312), 19:56:45/0, Flags: MJ                     

5 (10.96.112.20, 224.9.10.23) in v312 (tag e2/5), Uptime 19:59:50, Rate 51 (SM)

Source is directly connected. RP 1.1.1.1                   

Flags (0xb046ace1) SM SPT L2REG REGSUPP LSRC LRCV HW FAST TAG

fast ports: ethe 2/1                                       

AgeSltMsk: 00000002,  FID: 0x8188, MVID:    49 , AvgRate: 51, profile: none

Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 1        

      L3 (HW) 1:

          TR(e2/1,e2/1)(VL1), 17:29:50/182, Flags: IM            

Blocked OIF 1:                                             

e2/5(VL312), 19:56:59/0, Flags: MJ BI

----------

Step 4: Simulate increased network stress condition

STC 4/4 simulates IGMP report of 4000 groups, and STC 4/9 generates 800Mbps multicast traffic streams of the 4000 groups, but AR01.CH1 has IGMP snooping table only of 2048 groups since MLX support max 2048 groups for IGMP snooping.  Following is IGMP snooping configuration, and show command output of its snooping table in AR01.CH1:

----------

telnet@AR01.CH1#

vlan 212

tagged ethe 1/4 to 1/5

no spanning-tree

multicast active

telnet@AR01.CH1#sh ip multicast vlan 212

----------+-----+---------+---------------+-----+-----+------

VLAN State Mode      Active          Time (*, G)(S, G)

                           Querier         Query Count Count

----------+-----+---------+---------------+-----+-----+------

212 Ena   Active    Self            120 2048  0    

----------+-----+---------+---------------+-----+-----+------

Router ports:

Flags- R: Router Port,  V2|V3: IGMP Receiver,  P_G|P_SG: PIM Join

  1    (*, 226.0.7.202) 00:04:57        NumOIF: 1       profile: none

Outgoing Interfaces:

e1/4 vlan 212 ( V2) 00:04:57/128s

  2    (*, 226.0.3.229) 00:04:58        NumOIF: 1       profile: none

Outgoing Interfaces:

e1/4 vlan 212 ( V2) 00:04:58/130s

----------

 

Results

No impact to the application under this stress network environment of 6000 PIM mcache and IGMP snooping of 4000 groups. Following is CPU utilization of  management card and line cards, and ports statistics. Below shows no packet drop for all the STC streams:

 

STCTraffic_6000Entries.jpg

  STC Traffic Capture for 6000 PIM mcache (click to enlarge)

 

Below shows there is no CPU spike on management card & LP cards:

----------

telnet@AR01.CH1#sh cpu-utilization

... Usage average for all tasks in the last 1 seconds  ...

==========================================================

Name                    us/sec          %

idle                    955179          98

con                     15              0

mon                     98              0

flash                   47              0

dbg                     23              0

boot                    90              0

main                    0               0

itc                     0               0

tmr                     2139            0

ip_rx                   17716           1

telnet@AR01.CH1#sh cpu-utilization lp

SLOT  #:          LP CPU UTILIZATION in  %:

               in 1 second:  in 5 seconds: in 60 seconds: in 300 seconds:

     1:        1             1              1               1

     2:        1             1              1               1 

Below shows ports’ traffic statistics:

telnet@AR01.CH1_MLXe4_160233#sh statistics | in PO|Util

PORT 1/1 Counters:

    InUtilization               0.0%      OutUtilization             80.26%

PORT 1/2 Counters:

    InUtilization               0.0%      OutUtilization             80.26%

PORT 1/3 Counters:

    InUtilization               0.0%      OutUtilization             79.44%

PORT 1/4 Counters:

    InUtilization               0.0%      OutUtilization             80.26%

PORT 1/5 Counters:

InUtilization             80.26%      OutUtilization               0.0%

PORT 2/1 Counters:

    InUtilization             11.84%      OutUtilization              0.01%

PORT 2/2 Counters:

    InUtilization             11.83%      OutUtilization               0.0%

PORT 2/5 Counters:

    InUtilization              0.02%      OutUtilization              0.02%

----------

 

Test Case #4: MLXe Multicast Test using Informatica Messaging Application Test under Extreme Network Stress

 

DUT

This is the test configuration diagram.

 

TopologyWithMulticastTraffic.jpg

Figure 8.  Test Case #4 Configuration (click to enlarge)

 

Purpose

To measure the application’s behavior, PIM mcache table, and IP multicast resource allocation table under extreme network stress environment in that MLXs have more mcache entries than max supported number.

 

Test Procedure

Step 1: The same as step 1 in Test Case #1.

Step 2: The same as step 2 in Test Case #1.

Step 3: Simulate extreme network stress condition

STC 4/1 - 4/3 simulates IGMP report of 6,000 groups, and STC 4/5 - 4/7 generates            3 x 800Mbps multicast traffic streams for the 6,000 groups that creates 12000 PIM mcache entries of 6000 (*,G) and 6000 (S,G) entries each on AR01.CH1. However, approximately 10,000 mcache entries only are created on the MLX since the number of mcache entries exceeds max supported 8,000.

----------

telnet@AR01.CH1#sh ip igmp group

Total 6065 groups

--------------------------------------------------------------

Idx   Group Address    Port Intf   GrpCmpV Mode    Timer Srcs

----+----------------+------+------+-------+-------+-----+----

   1 224.9.10.23      e2/5 v312      Ver2 exclude   239 0

   2 224.10.10.21     e2/5 v312      Ver2 exclude   238 0

   3 224.23.10.2      e2/5 v312      Ver2 exclude   239 0

telnet@AR01.CH1#sh ip pim mcache          

IP Multicast Mcache Table

Entry Flags : SM  - Sparse Mode, SSM - Source Specific Mutlicast, DM - Dense Mode

RPT    - RPT Bit, SPT - SPT Bit, LSRC - Local Source, LRCV - Local Receiver

HW - HW Forwarding Enabled, FAST - Resource Allocated, TAG - Need For Replication Entry

REGPROB - Register In Progress, REGSUPP - Register Suppression Timer

MSDPADV - Advertise MSDP, NEEDRTE - Route Required for Src/RP,  PRUN - DM Prune Upstream

Interface Flags: IM - Immediate, IH - Inherited, WA - Won Assert

MJ - Membership Join, MI - Membership Include, ME - Membership Exclude

BR - Blocked RPT, BA - Blocked Assert, BF - Blocked Filter, BI - Blocked IIF

Total entries in mcache: 10568

1     (*, 224.9.10.23) RP 1.1.1.1, in e2/1 (e2/1), Uptime 20:32:10 (SM)

upstream neighbor 11.96.0.1

Flags (0x002604a0) SM RPT LRCV TAG

slow ports: ethe 2/5

AgeSltMsk: 00000000,  FID: NotReq, MVID: NotReq, profile: none

Forwarding_oif: 1, Immediate_oif: 1, Blocked_oif: 0

      L3 (SW) 1:

e2/5(VL312), 20:30:56/0, Flags: MJ

2 (10.75.112.20, 224.9.10.23) in e2/1 (e2/1), Uptime 20:33:47, Rate 32 (SM)

upstream neighbor 11.96.0.1

      Flags (0x700680e1) SM SPT LRCV HW FAST TAG

fast ports: ethe 2/5                                       

AgeSltMsk: 00000002,  FID: 0x8343, MVID:    28 , AvgRate: 32, profile: none

Forwarding_oif: 1, Immediate_oif: 0, Blocked_oif: 0        

      L3 (HW) 1:

e2/5(VL312), 20:30:56/0, Flags: MJ

----------

Step 4: Simulate increase in extreme network stress condition

STC 4/4 simulates IGMP report of 4000 groups, and STC 4/9 generates 800Mbps multicast traffic streams of the 4000 groups, but IGMP snooping table of AR01.CH1 has 2048 groups only as shown in the output of “show ip multicast vlan 212” below since MLX supports max 2048 groups per system in IGMP snooping. Following is IGMP snooping configuration, and show command output of its snooping table in AR01.CH1:

----------

telnet@AR01.CH1#

vlan 212

tagged ethe 1/4 to 1/5

no spanning-tree

multicast active

telnet@AR01.CH1#show ip multicast vlan 212

----------+-----+---------+---------------+-----+-----+------

VLAN State Mode      Active          Time (*, G)(S, G)

                           Querier         Query Count Count

----------+-----+---------+---------------+-----+-----+------

212 Ena   Active    Self            120   2048  0

----------+-----+---------+---------------+-----+-----+------

Router ports:

Flags- R: Router Port,  V2|V3: IGMP Receiver,  P_G|P_SG: PIM Join

  1    (*, 226.0.7.202) 00:04:57        NumOIF: 1       profile: none

Outgoing Interfaces:

e1/4 vlan 212 ( V2) 00:04:57/128s

  2    (*, 226.0.3.229) 00:04:58        NumOIF: 1       profile: none

Outgoing Interfaces:

e1/4 vlan 212 ( V2) 00:04:58/130s

-----------

Results

1. Approximately 40% packets loss on STC and CPU hike of 24% on the management card of AR01.CH1 as shown below under this extremely stressed network, but it is expected behavior since there are more mache entries than max supported:

---------

telnet@AR01.CH1# sh cpu

... Usage average for all tasks in the last 1 seconds  ...

==========================================================

Name                    us/sec          %

idle                    737986          76

ip_rx                   25403           2

mcast                   222473          22                       

telnet@AR01.CH1# sh cpu lp

SLOT #:          LP CPU UTILIZATION in  %:

in 1 second:  in 5 seconds:  in 60 seconds: in 300 seconds:

1:        2             3              2               1 

2:        1             2              2               2 

----------

STCTraffic_12000Entries.jpg

   STC Traffic Capture for 12,000 PIM mcache (click to enlarge)

2. No obvious packet loss was observed from application’s perspective.              

Contributors