xCAT/xcat-core
2
0
Fork 0
mirror of https://github.com/xcat2/xcat-core.git synced 2024-11-21 09:02:00 +00:00
Code Issues Releases Wiki Activity
Page: XCAT_IPv6_support_on_Linux
Pages
2014_10_01_Bash_vulnerability 2014_11_04_Poodle_Attack AIX_NIM_NFSv4_support Add_Extra_Parameters_for_confignics_Postscript CRHS like_function CRHS like_function_enhancements CSM_CFM like_function_to_sync_config_files_to_the_nodes Capture_Linux_Image_from_running_node_to_create_a_stateless statelite_image Chef_databag_support_for_deploying_openstack_in_xCAT Cluster_config_file Conserver_Replacement Coordinated_Cluster_Shutdown_and_Bringup Copycds_enhancement Coral PDU supports Create_osimage_definition_for_Management_Node Cumulus_OS_Upgrade_Support_Test DFM_Service_Node_Hierarchy_support DHCPv6_management DNS_Enhancments DNS_master_slave_support Def_commands_to_display set_nic_attributes_more_easily Deploying_OpenStack_Baremetal_Nodes_Using_xCAT Disable_ssh_root_passwordless_node to node_access Document_large_cluster_tunables,_etc. Dynamic_node_groups_support Energy_Management_for_Flex Energy_management_support_for_blade Enhance_NTP_setup_support Enhancement_of_lsslp_ flexdiscover Enhancement_on_makeroutes_and_snmove_commands Exporting_more_table_attributes_to_nodes Filling_the_osimage_attributes_with_the_files_that_are_found_in_the_search_path Full_Automation_of_DB2_Setup_on_Service_Nodes Function Verification Test Design of xCAT High Availability GPFS_statelite_nodes Getmacs_and_rnetboot_enhancement Getnodetype_performance_imprvoments_ _phase_I HA_NFS_on_AIX_service_nodes HA_Service_Nodes_for_AIX HA_Service_Nodes_for_AIX_ _using_External_NFS_Server HFI_MAC_address_failover HPC_Integration_Phase_1 Hardware_Inventory_for_Node Health_Check_Script_Framework Hierarchical_Design Home Image_Provisioning Image_import_and_export_enhancement_for_Linux Improve_perl_array_search_efficiency_for_scalability Improvement_of_pre creating_mypostscripts_on_linux Initrd_Enhancement_separate_initrd_hacking_from_noteset_get_drivers_from_os_update_distro Kdump_for_Linux_diskless_nodes Linux_AIX_mixed_Clusters_postscripts_cleanup Local_Disk_support_for_statelite Lstree_for_Servicenode_Hardware_and_VM_hierarchies Management_Node_Pool Management_Node_Setup_Verification Managing_Multiple_Sites_and_Clusters Mellanox_Switch_Support Mini design for openbmc python framework Mini design_of_updating_pre defined_groups_for_nodes Minimize_the_install setup_steps_of_xcat_on_the_MN Mkvlan_enhancements Multiple_Zone_Support2 Multiple_pkgdir_support_with_full_installation NFS_Redundancy_Investigation NFS_redundancy New_Cloud_Tables_in_xCAT Node_Status_and_Application_Status Node_definition_installation_service_node_setup_verification Nodeset_defaults_when_node_provmethod=osimage_is_set OS_Deployment_Monitoring_And_Automatic_Retry OS_Distro_Update_xCAT_command_osdistroupdate OpenStack Chef Cookbook_xCAT_Integration OpenStack_Deployment_using_Chef OpenStack_Deployment_using_Puppet Ospkgs_enhancements_to_install upgrade_OS_packages_from_Updates Overview_of_HA_Services_for_Diskless_AIX_Nodes Power8_BE_support Predict_long_nicname_for_RH7 Proposal_for_docker_Support_in_xCAT RHEL7_Support RHEV_support Rcons_hierachy_support Redfish_support Refine_the_usage_of_installnic_and_primarynic_attributes Release processs automation investigation Resolving_Nodenames_for_System_p_HW_Ctrl Run_parallel_commands_for_Ethernet_switches Run_postbootscripts_after_each_diskful_reboot SLES12_Support SLES_11_SP1_SDK_support Secure root support Sequential_Discovery Sequential_Discovery_phase_2 Service_Node_Scenarios_and_Retry_Logic_in_Postscript_Process Service_node_manual_takeover_on_AIX Service_node_take_over Setup_IP_routes_between_the_MN_and_the_CNs Shared data based xCAT MN HA mini design Shared data based xCAT MN HA user case Shared data based xCAT MN HA Specify_the_disk_partitions_during_diskfull_install Steps to install xcat_openbmc_py3 on RH8 Support uploading openbmc firmware in parallel Support_Management_Node_in_the_servicenode_table Support_for_Redundant_BPAs_and_FSPs Support_multiple_hardware_control_points Support_osimage.pkgdir_with_multiple_paths_in_genimage_and_diskfull_installation Support_postscripts_associated_with_rsync_files_to_be_run_after_rsync_in_xdcp Support_to_inject_drivers_from_Driver_RPM_Package Support_xcatd_running_user_defined_sql_scripts_on_start_and_foreign_keys Supporting_driver_update_disk_for_Linux Supporting_the_scalable_functions_for_HX5_blade Switch_Discovery_And_Switch_Management System_p_Discovery_Enhancements System_z_hierarchical_tree Test Design of HA Test Design of TestCase Label Test Design of `goconserver` Regression Testing Test Design of `goconserver` Support Test Design of configib Test Design of cumulus osimage Test Design of cumulus_osimage Test Design of gonconserver redirect logs Test Design of hardware control commands regression Test Design of makegocons option l Test Design of rbeacon python version Test Design of rcons based goconserver Test Design of revenlog python version Test Design of rflash python version Test Design of rinv python version Test Design of rpower python version Test Design of rsetboot python version Test Design of rspconfig python version Test Design of rvitals python version Test Design of syncfiles Test Design of xcat inventory diff Test Design of xcat inventory export import performance Test Design of xcat inventory option e Test Design of xcat inventory osimage environment variables export part Test Design of xcat inventory osimage environment variables Test Design of xcat inventory osimage phase2 Test Design of xcat inventory osimage support share file directory Test Design of xcat inventory osimage Test Design of xcat inventory phase 1 Test Design of xcat inventory regular expression Test Design of xcat inventory save keys Test Design of xcat inventory user case Test Design of xcat terraform provider phase1 Test Report of `goconserver` Test cases against nodeset error handling Test_Case_Designs_FVT The mini design of osimage version control(initial draft) The mini design of syncfiles for cumulus The mini design of using xcat inventory templates to setup xCAT cluster with ansible The_remote_console_support_of_hmc Trace_Facility_for_xCAT Ubuntu_supports_internet_repository Updatenode_Performance_Enhancements Updatenode_Sync_Service_Nodes_for_install Updatenode_enhancement_for_security_keys_and_certificates Updatenode_support_to_run_mypostscript_sections Updatenode_to_the_Management_Node Updates_to_xcatsetup_and_Cluster_Config_File Windows_Multiple_Disks_Partitions_Configuration Windows_Multiple_Nics_Configuration Windows_Multiple_Winpes Windows_Support_to_Run_Postscripts_Postbootscripts_in_Windows_Deployment Wish_List_for_xCAT_2 Wish_List_for_xCAT_FVT XCAT 2 Mini Designs for New Features XCAT_2.10_Release_Notes XCAT_2.11.1_Release_Notes XCAT_2.11_ESP_Release_Notes XCAT_2.11_Release_Notes XCAT_2.12.1_Release_Notes XCAT_2.12.2_Release_Notes XCAT_2.12.3_Release_Notes XCAT_2.12.4_Release_Notes XCAT_2.12.5_Release_Notes XCAT_2.12_Release_Notes XCAT_2.13.10_Release_Notes XCAT_2.13.11_Release_Notes XCAT_2.13.1_Release_Notes XCAT_2.13.2_Release_Notes XCAT_2.13.3_Release_Notes XCAT_2.13.4_Release_Notes XCAT_2.13.5_Release_Notes XCAT_2.13.6_Release_Notes XCAT_2.13.7_Release_Notes XCAT_2.13.8_Release_Notes XCAT_2.13.9_Release_Notes XCAT_2.13_Release_Notes XCAT_2.14.1_Release_Notes XCAT_2.14.2_Release_Notes XCAT_2.14.3_Release_Notes XCAT_2.14.4_Release_Notes XCAT_2.14.5_Release_Notes XCAT_2.14.6_Release_Notes XCAT_2.14_Release_Notes XCAT_2.15.1_Release_Notes XCAT_2.15_Release_Notes XCAT_2.16.1_Release_Notes XCAT_2.16.2_Release_Notes XCAT_2.16.3_Release_Notes XCAT_2.16.4_Release_Notes XCAT_2.16.5_Release_Notes XCAT_2.16_Release_Notes XCAT_2.17_Release_Notes XCAT_2.9.2_Release_Notes XCAT_2.9.2_aix_Release_Notes XCAT_2.9.3_Release_Notes XCAT_2.9.4_Release_Notes XCAT_Cluster_in_SoftLayer XCAT_HPC_Integration_Files_per_OS,_Architecture,_Product XCAT_Health_Check XCAT_IPv6_support_on_AIX XCAT_IPv6_support_on_Linux XCAT_Mini Design_Template XCAT_Monitoring_Failure_Analysis_Conditions_Batched_Events XCAT_Postscripts_framework_enhancements XCAT_Provisioning_OS_on_consistent_disk XCAT_REST_API_enhancements XCAT_Support_for_Kits_and_Kit_Components XCAT_probe XCAT_support_for_multiple_network_domains Xcatexport_and_xcatimport_commands Xcatmon_and_appstatus_for_HPC_products Xeon_Phi_support cofignetwork_for_RH7 confignetwork integrate confignics_configeth_configib functions usage determine_install_disk diskdiscover_and_configraid docker_instances_lifecycle_management enhancement_for_rinstall_command logging pre script to make debug easy makedhcp_append_statement_automatically_rather_than_get_statement_from_nodeset mini design of ZTP based onie switch discovery and configuration minidesign Switch based switch discovery minidesign_xcatprobe node status enhancement ntp_enhancement openbmc_support post.xcat_restructure rflash_streamline_procedure rscan command for hypervisor rspconfig network rspconfig ntpservers rspconfig password scaleup_basedon_xcat_in_openshift_cluster supporting_partitionfile_for_Ubuntu supporting_persistent_kernel_cmdline_after_diskful_installation test_sidebar the mini design of xcat inventory export and import for osimage use fifo pipe to enhance getadapter xCAT Build Schedule xCAT_2.12_ReleaseInformation xCAT_2.13_ReleaseInformation xCAT_2.14_ReleaseInformation xCAT_2.15_ReleaseInformation xCAT_Docker_Image xcatprobe_framework xcatprobe_requests
16 XCAT_IPv6_support_on_Linux
ligc edited this page 2015-07-30 04:22:49 -04:00
Table of Contents

Table of Contents

Design_Warning

Note: this design is only for Linux, IPv6 support on AIX will be covered by a separate design.

Overview

This item icludes only the "black box" model, which means:

  • The os provisioning is done through ipv4 network.
  • The ethernet adapters and Infiniband adapters on the compute nodes could be configured with ipv4 only, or ipv6 only, or both
  • The compute nodes could use ipv6 default gateway to communicate with the external ipv6 nodes.
  • The compute nodes could be configured to communicate with different ipv6 subnets through different gateway.
  • The management node should have ipv6 forwarding enabled in case the compute nodes will use the management node as the ipv6 gateway.

The specific support xcat needs to provide:

  • Use makehosts command to setup IPv6 hostnames in /etc/hosts
  • Use confignics postscript to configure the IPv6 addresses on the compute nodes, for both Ethernet and Infiniband
  • Use confignics postscript to configure IPv6 gateway on the compute nodes, for both Ethernet and Infiniband
  • Use makeroutes to configure IPv6 routing on the compute nodes, for both Ethernet and Infiniband
  • xcatconfig will configure IPv6 forwarding if there is any non-link-local IPv6 network setup on the management node.

IPv6 support with Ubuntu is not covered by this doc, it will be addressed in a separate item

Prerequisite

To configure IPv6 environment, the perl IPv6 packages need to be installed on the management node:

 yum install perl-IO-Socket-INET6 perl-IO-Socket-SSL perl-Socket6
 service xcatd stop
 service xcatd start

Configure IPv6 addresses and gateway on the compute nodes for Ethernet adapters

1. Plan the IPv6 addresses for the compute nodes

For the IPv6 addresses allocation, there are two choices: either defining the ip addresses in the nics table manually, or use the DHCPv6/RA to assign dynamic IPv6 addresses and use ddns to map the hostnames and nic. Setting up the DHCPv6/RA to assign IPv6 addresses is beyond the scope of this doc, this doc will only cover the first scenario, i.e., defining the ip addresses in the nics table manually.

2. Specify the node ip addresses and gateway configuration in nics table.

Here is an example on how to configure the IPv4/IPv6 addresses for the Ethernet adapters in the nics table.

 [root@ls21n01 ~]# tabdump nics
 #node,nicips,nichostnamesuffixes,nictypes,niccustomscripts,nicnetworks,nicaliases,comments,disable
 "ipv6cn1","eth0!10.1.89.7|fd56::214:5eff:fe15:849b|2000::214:5eff:fe15:849b,eth1!11.1.89.7|fd57::214:5eff:fe15:849b|2001::214:5eff:fe15:849b,eth2!12.1.89.7|fd58::214:5eff:fe15:849b|2002::214:5eff:fe15:849b","eth0!|-eth0-ipv6-1|-eth0-ipv6-2,eth1!-eth1|-eth1-ipv6-1|-eth1-ipv6-2,eth2!-eth2|-eth2-ipv6-1|-eth2-ipv6-2","eth0!Ethernet,eth1!Ethernet,eth2!Ethernet",,"eth0!10_1_0_0-255_255_0_0|fd56::/64|2000::/64,eth1!11_1_0_0-255_255_0_0|fd57::/64|2001::/64,eth2!12_1_0_0-255_255_0_0|fd58::/64|2002::/64",,,
   [root@ls21n01 ~]#

3. Add network entries for the nics in the networks table

 [root@ls21n01 ~]# tabdump networks
 #netname,net,mask,mgtifname,gateway,dhcpserver,tftpserver,nameservers,ntpservers,logservers,dynamicrange,staticrange,staticrangeincrement,nodehostname,ddnsdomain,vlanid,domain,comments,disable
 "10_1_0_0-255_255_0_0","10.1.0.0","255.255.0.0","eth1","<xcatmaster>",,"10.1.0.218",,,,,,,,,,"clusters.com",,
 "11_1_0_0-255_255_0_0","11.1.0.0","255.255.0.0","eth1",,,,,,,,,,,,,"clusters.com",,
 "12_1_0_0-255_255_0_0","12.1.0.0","255.255.0.0","eth1",,,,,,,,,,,,,"clusters.com",,
 "fd56::/64","fd56::/64","/64","eth0","fd56::214:5eff:fe15:1",,,,,,,,,,,,"clusters.com",,
 "fd57::/64","fd57::/64","/64","eth1",,,,,,,,,,,,,"clusters.com",,
 "fd58::/64","fd58::/64","/64","eth2",,,,,,,,,,,,,"clusters.com",,
 "2000::/64","2000::/64","/64","eth0",,,,,,,,,,,,,"clusters.com",,
 "2001::/64","2001::/64","/64","eth1",,,,,,,,,,,,,"clusters.com",,
 "2002::/64","2002::/64","/64","eth2",,,,,,,,,,,,,"clusters.com",,
 [root@ls21n01 ~]#

In this example, the fd56::214:5eff:fe15:1 is used as the IPv6 default gateway for the compute node ipv6cn1. Please be aware that only one IPv6 default gateway could be specified, if you would like to setup different gateways for different IPv6 subnets, see the section "Configure IPv6 routing on the compute nodes" for more details.

4. Setup /etc/hosts for the IPv6 entries

The makehosts is able to setup both the IPv4 hostnames and IPv6 hostnames in /etc/hosts based on information stored in the nics table, here is an example:

4.1. Run makehosts to setup the IPv6 entries in /etc/hosts

 makehosts ipv6cn1

4.2. Check the IPv6 entries are setup correctly in /etc/hosts

 [root@ls21n01 ~]# cat /etc/hosts | grep ipv6cn1
 12.1.89.7 ipv6cn1-eth2 ipv6cn1-eth2.clusters.com  
 fd58::214:5eff:fe15:849b ipv6cn1-eth2-ipv6-1 ipv6cn1-eth2-ipv6-1.clusters.com  
 2002::214:5eff:fe15:849b ipv6cn1-eth2-ipv6-2 ipv6cn1-eth2-ipv6-2.clusters.com  
 11.1.89.7 ipv6cn1-eth1 ipv6cn1-eth1.clusters.com  
 fd57::214:5eff:fe15:849b ipv6cn1-eth1-ipv6-1 ipv6cn1-eth1-ipv6-1.clusters.com  
 2001::214:5eff:fe15:849b ipv6cn1-eth1-ipv6-2 ipv6cn1-eth1-ipv6-2.clusters.com  
 10.1.89.7 ipv6cn1 ipv6cn1.clusters.com  
 fd56::214:5eff:fe15:849b ipv6cn1-eth0-ipv6-1 ipv6cn1-eth0-ipv6-1.clusters.com  
 2000::214:5eff:fe15:849b ipv6cn1-eth0-ipv6-2 ipv6cn1-eth0-ipv6-2.clusters.com  
 [root@ls21n01 ~]#

5. Configure IPv4/IPv6 addresses and IPv6 default gateway on the compute node

The postscript confignics can configure the IPv4 and IPv6 addresses and gateway on the compute nodes. It could be called through whatever possible ways:

5.1 Setup the IPv4/IPv6 addresses and IPv6 default gateway on the compute node during operating system provisioning

 chdef ipv6cn1 -p postscripts=confignics
 or
 chdef ipv6cn1 -p postscripts="confignics -s"

The "-s" flag specified with confignics indicates that the installation nic should be setup as static configuration using the information in the nics table.

 nodeset ipv6cn1 osimage=xxx
 rpower ipv6cn1 reset

5.2 Setup the IPv4/IPv6 addresses and IPv6 default gateway on the compute node through updatenode

 updatenode ipv6cn1 -P confignics
 or
 updatenode ipv6cn1 -P "confignics -s"

The "-s" flag specified with confignics indicates that the installation nic should be setup as static configuration using the information in the nics table.

6. Verify the configuration on the compute nodes

When the confignics finishes setup the IPv4 and IPv6 addresses and gateway on the compute nodes, the following commands could be used to verify the configuration:

 [root@ipv6cn1 ~]# ifconfig
 eth0    Link encap:Ethernet  HWaddr 42:C6:0A:01:59:07  
         inet addr:10.1.89.7  Bcast:10.1.255.255  Mask:255.255.0.0
         inet6 addr: 2000::214:5eff:fe15:849b/64 Scope:Global
         inet6 addr: fd56::214:5eff:fe15:849b/64 Scope:Global
         inet6 addr: fe80::40c6:aff:fe01:5907/64 Scope:Link
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:14663 errors:0 dropped:0 overruns:0 frame:0
         TX packets:959 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:1000 
         RX bytes:1053278 (1.0 MiB)  TX bytes:130254 (127.2 KiB)
 
 eth1    Link encap:Ethernet  HWaddr 42:56:0A:01:59:07  
         inet addr:11.1.89.7  Bcast:11.1.255.255  Mask:255.255.0.0
         inet6 addr: fd55:faaf:e1ab:336:4056:aff:fe01:5907/64 Scope:Global
         inet6 addr: fe80::4056:aff:fe01:5907/64 Scope:Link
         inet6 addr: 2001::214:5eff:fe15:849b/64 Scope:Global
         inet6 addr: fd57::214:5eff:fe15:849b/64 Scope:Global
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:12833 errors:0 dropped:0 overruns:0 frame:0
         TX packets:21 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:1000 
         RX bytes:824536 (805.2 KiB)  TX bytes:1642 (1.6 KiB)
 
 eth2    Link encap:Ethernet  HWaddr 42:0D:0A:01:59:07  
         inet addr:12.1.89.7  Bcast:12.1.255.255  Mask:255.255.0.0
         inet6 addr: 2002::214:5eff:fe15:849b/64 Scope:Global
         inet6 addr: fd58::214:5eff:fe15:849b/64 Scope:Global
         inet6 addr: fe80::400d:aff:fe01:5907/64 Scope:Link
         UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
         RX packets:34 errors:0 dropped:0 overruns:0 frame:0
         TX packets:17 errors:0 dropped:0 overruns:0 carrier:0
         collisions:0 txqueuelen:1000 
         RX bytes:3696 (3.6 KiB)  TX bytes:1226 (1.1 KiB)
 [root@ipv6cn1 ~]# 





 [root@ipv6cn1 ~]# ip -6 route show  default
 default via fd56::214:5eff:fe15:1 dev eth0  metric 1  mtu 1500 advmss 1440 hoplimit 4294967295
 default via fe80::226:88ff:fe57:b7f0 dev eth1  proto kernel  metric 1024  expires 0sec mtu 1500 advmss 1440 hoplimit 64
 [root@ipv6cn1 ~]#

Configure IPv6 addresses on the compute nodes for Ethernet adapters

Here is an example on how to configure the IPv4/IPv6 addresses for the Infiniband adapters in the nics table.

 [root@ls21n01 ~]# tabdump nics
 "dx360m3n06","ib0!11.1.89.10|21.1.89.10|fd57::214:5eff:fe15:8496|2000::214:5eff:fe15:8496,ib1!12.1.89.10|22.1.89.10|fd58::214:5eff:fe15:8496|2001::214:5eff:fe15:8496","ib0!-ib0|-ib0-2|-ib0-ipv6-1|-ib0-ipv6-2,ib1!-ib1|-ib1-2|-ib1-ipv6-1|-ib1-ipv6-2","ib0!Infiniband,ib1!Infiniband",,"ib0!11_1_0_0-255_255_0_0|21_1_0_0-255_255_0_0|fd57::/64|2000::/64,ib1!12_1_0_0-255_255_0_0|22_1_0_0-255_255_0_0|fd58::/64|2001::/64",,,
 [root@ls21n01 ~]#

3. Add network entries for the nics in the networks table

 [root@ls21n01 ~]# tabdump networks
 #netname,net,mask,mgtifname,gateway,dhcpserver,tftpserver,nameservers,ntpservers,logservers,dynamicrange,staticrange,staticrangeincrement,nodehostname,ddnsdomain,vlanid,domain,comments,disable
 "11_1_0_0-255_255_0_0","11.1.0.0","255.255.0.0","ib0",,,,,,,,,,,,,"clusters.com",,
 "21_1_0_0-255_255_0_0","21.1.0.0","255.255.0.0","ib0",,,,,,,,,,,,,"clusters.com",,
 "12_1_0_0-255_255_0_0","12.1.0.0","255.255.0.0","ib1",,,,,,,,,,,,,"clusters.com",,
 "22_1_0_0-255_255_0_0","22.1.0.0","255.255.0.0","ib1",,,,,,,,,,,,,"clusters.com",,
 "fd57::/64","fd57::/64","/64","ib0",,,,,,,,,,,,,"clusters.com",,
 "fd58::/64","fd58::/64","/64","ib1",,,,,,,,,,,,,"clusters.com",,
 "2000::/64","2000::/64","/64","ib0",,,,,,,,,,,,,"clusters.com",,
 "2001::/64","2001::/64","/64","ib1",,,,,,,,,,,,,"clusters.com",,
 [root@ls21n01 ~]#

4. Setup /etc/hosts for the Infiniband IPv6 entries

The makehosts is able to setup both the IPv4 hostnames and IPv6 hostnames in /etc/hosts based on information stored in the nics table, here is an example:

4.1. Run makehosts to setup the IPv6 entries in /etc/hosts

 makehosts dx360m3n06

4.2. Check the IPv6 entries are setup correctly in /etc/hosts

 [root@ls21n01 ~]# cat /etc/hosts | grep dx360m3n06
 10.1.0.236 dx360m3n06.clusters.com dx360m3n06
 12.1.89.10 dx360m3n06-ib2 dx360m3n06-ib2.clusters.com 
 22.1.89.10 dx360m3n06-ib2-2 dx360m3n06-ib2-2.clusters.com 
 fd58::214:5eff:fe15:8496 dx360m3n06-ib2-ipv6-1 dx360m3n06-ib2-ipv6-1.clusters.com 
 2001::214:5eff:fe15:8496 dx360m3n06-ib2-ipv6-2 dx360m3n06-ib2-ipv6-2.clusters.com 
 11.1.89.10 dx360m3n06-ib1 dx360m3n06-ib1.clusters.com 
 21.1.89.10 dx360m3n06-ib1-2 dx360m3n06-ib1-2.clusters.com 
 fd57::214:5eff:fe15:8496 dx360m3n06-ib1-ipv6-1 dx360m3n06-ib1-ipv6-1.clusters.com 
 2000::214:5eff:fe15:8496 dx360m3n06-ib1-ipv6-2 dx360m3n06-ib1-ipv6-2.clusters.com 
 [root@ls21n01 ~]#

5. Configure IPv4/IPv6 addresses for the Infiniband adapters

The postscript confignics can configure the IPv4 and IPv6 addresses for the Infiniband adapters. It could be called through whatever possible ways:

5.1 Setup the IPv4/IPv6 addresses for the Infiniband adapters during operating system provisioning

 chdef dx360m3n06 -p postscripts=confignics


 nodeset ipv6cn1 osimage=xxx
 rpower ipv6cn1 reset

5.2 Setup the IPv4/IPv6 addresses and IPv6 default gateway on the compute node through updatenode

 updatenode ipv6cn1 -P confignics

6. Verify the configuration on the compute nodes

When the confignics finishes setup the IPv4 and IPv6 addresses for the Infiniband adapters, the following commands could be used to verify the configuration:

 [root@dx360m3n06 ~]# ip addr show dev ib0
 9: ib0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 65520 qdisc pfifo_fast state UP qlen 1024
   link/infiniband 80:00:00:48:fe:80:00:00:00:00:00:01:00:02:c9:03:00:4e:47:cf brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
   inet 11.1.89.10/16 brd 11.1.255.255 scope global ib0
   inet 21.1.89.10/16 brd 21.1.255.255 scope global ib0:2
   inet6 2000::214:5eff:fe15:8496/64 scope global 
      valid_lft forever preferred_lft forever
   inet6 fd57::214:5eff:fe15:8496/64 scope global 
      valid_lft forever preferred_lft forever
   inet6 fe80::202:c903:4e:47cf/64 scope link 
      valid_lft forever preferred_lft forever
 [root@dx360m3n06 ~]# ip addr show dev ib1
 10: ib1: <BROADCAST,MULTICAST,UP> mtu 65520 qdisc pfifo_fast state UNKNOWN qlen 1024
   link/infiniband 80:00:00:49:fe:80:00:00:00:00:00:00:00:02:c9:03:00:4e:47:d0 brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
   inet 12.1.89.10/16 brd 12.1.255.255 scope global ib1
   inet 22.1.89.10/16 brd 22.1.255.255 scope global ib1:2
   inet6 2001::214:5eff:fe15:8496/64 scope global tentative 
      valid_lft forever preferred_lft forever
   inet6 fd58::214:5eff:fe15:8496/64 scope global tentative 
      valid_lft forever preferred_lft forever
 [root@dx360m3n06 ~]#

Note: this item only covers the IPoIB IPv6 configuration on the compute nodes from the IP layer perspective, the IPv6 in IPoIB configuration depends on the Infiniband IPv6 support structure, like Infiniband switches, operating systems and device drivers, you might need to do more configuration for IPv6 work in IPoIB environment, or even worse, it is possible the IPv6 could not work with some specific IPoIB configuration.

Configure IPv6 routing on the compute nodes

The IPv6 routing setup on the compute nodes follows the same procedure as the IPv6 routing setup.

1. Add the network routes in the routes table

 [root@ls21n01 ~]# tabdump routes
 #routename,net,mask,gateway,ifname,comments,disable
 "13route","13.1.0.0","255.255.0.0","11.1.89.1",,,
 "14route","14.1.0.0","255.255.0.0","12.1.89.1",,,
 "fd59route","fd59::/64","/64","fd56::214:5eff:fe15:1","eth0",,
 [root@ls21n01 ~]#

2. Associate the routes to the compute nodes

 [root@ls21n01 ~]# chdef ipv6cn1 routenames=13route,14route,fd59route
 1 object definitions have been created or modified.
 [root@ls21n01 ~]#

3. Setup up the routing

To setup the routes on the management node:

 makeroutes -r 13route,14route,fd59route

To setup the routes on the compute nodes during operating system provisioning:

 chdef ipv6cn1 -p postscripts=setroute
 nodeset ipv6cn1 osimage
 rpower ipv6cn1 reset

To setup the routes on the compute nodes when the compute nodes are up and running:

 makeroutes ipv6cn1 -r 13route,14route,fd59route

To remove the routes on the management node:

 makeroutes -d -r 13route,14route,fd59route

To remove the routes on the compute nodes:

 makeroutes ipv6cn1 -d -r 13route,14route,fd59route

4. Verify the routing setup

When the makeroutes finishes the routes setup on the management node or compute nodes, the following commands could be used to verify the routes setup.

 [root@ipv6cn1 ~]# ip route show
 14.1.0.0/16 via 12.1.89.1 dev eth2 
 12.1.0.0/16 dev eth2  proto kernel  scope link  src 12.1.89.7 
 10.1.0.0/16 dev eth0  proto kernel  scope link  src 10.1.89.7 
 13.1.0.0/16 via 11.1.89.1 dev eth1 
 169.254.0.0/16 dev eth0  scope link  metric 1002 
 169.254.0.0/16 dev eth1  scope link  metric 1003 
 169.254.0.0/16 dev eth2  scope link  metric 1004 
 11.1.0.0/16 dev eth1  proto kernel  scope link  src 11.1.89.7 
 [root@ipv6cn1 ~]# 


 [root@ipv6cn1 ~]# ip -6 route show fd59::/64
 fd59::/64 via fd56::214:5eff:fe15:1 dev eth0  metric 1024  mtu 1500 advmss 1440 hoplimit 4294967295
 [root@ipv6cn1 ~]#

On RedHat:

 [root@ipv6cn1 ~]# cat /etc/sysconfig/static-routes
 # xCAT_CONFIG_START
 any net 13.1.0.0 netmask 255.255.0.0 gw 11.1.89.1 
 any net 14.1.0.0 netmask 255.255.0.0 gw 12.1.89.1 
 # xCAT_CONFIG_END
 [root@ipv6cn1 ~]# cat /etc/sysconfig/static-routes-ipv6 
 # xCAT_CONFIG_START
 eth0 fd59::/64 fd56::214:5eff:fe15:1
 # xCAT_CONFIG_END
 [root@ipv6cn1 ~]#

On SLES:

 ipv6cn2:~ # cat /etc/sysconfig/network/routes 
 default fd56::214:5eff:fe15:1 - -
 # xCAT_CONFIG_START
 13.1.0.0 11.1.89.1 255.255.0.0
 14.1.0.0 12.1.89.1 255.255.0.0 
 fd59::/64 fd56::214:5eff:fe15:1 - -
 # xCAT_CONFIG_END
 ipv6cn2:~ #

Setup the ipforward for IPv6 on MN

If there is any IPv6 network entry is defined in the networks table, when the xCAT is installed on the management node, the IPv6 forwarding will be enabled.

To check the IPv6 forwarding on the management node:

 [root@ls21n01 ~]# cat /etc/sysctl.conf | grep net.ipv6.conf.all.forwarding
 net.ipv6.conf.all.forwarding = 1
 [root@ls21n01 ~]# cat /proc/sys/net/ipv6/conf/all/forwarding
 1
 [root@ls21n01 ~]#

Note: this is a temporary solution, in some futher xCAT release, the network services setup on the management node will be based on the information in the servicenode table.

Other Design Considerations

   *Required approvers:  Bruce Potter
   *Affect on other components:  confignics(configeth/configib) and makeroutes for IPv4
   *External interface changes, documentation, and usability issues: use the existing confignics and makeroutes command. Documentation at [Configuring_IPv6_in_Cluster].
   *Packaging, installation, dependencies:  N/A
   *Portability and platforms (HW/SW) supported:  N/A
   *Performance and scaling considerations:  N/A
   *Migration and coexistence:  N/A
   *Serviceability: N/A
   *Security:  N/A
   *NLS and accessibility:  N/A
   *Invention protection:  N/A

News

History

  • Oct 22, 2010: xCAT 2.5 released.
  • Apr 30, 2010: xCAT 2.4 is released.
  • Oct 31, 2009: xCAT 2.3 released.
    xCAT's 10 year anniversary!
  • Apr 16, 2009: xCAT 2.2 released.
  • Oct 31, 2008: xCAT 2.1 released.
  • Sep 12, 2008: Support for xCAT 2
    can now be purchased!
  • June 9, 2008: xCAT breaths life into
    (at the time) the fastest
    supercomputer on the planet
  • May 30, 2008: xCAT 2.0 for Linux
    officially released!
  • Oct 31, 2007: IBM open sources
    xCAT 2.0 to allow collaboration
    among all of the xCAT users.
  • Oct 31, 1999: xCAT 1.0 is born!
    xCAT started out as a project in
    IBM developed by Egan Ford. It
    was quickly adopted by customers
    and IBM manufacturing sites to
    rapidly deploy clusters.