How to Validate Network and Name Resolution Setup for the Clusterware and RAC (Doc ID 1054902.1)

APPLIES TO:

Oracle Database - Enterprise Edition - Version 10.1.0.2 to 11.2.0.3 [Release 10.1 to 11.2]
Information in this document applies to any platform.

PURPOSE

Cluster Verification Utility (aka CVU, command runcluvfy.sh or cluvfy) does very good checking on the network and name resolution setup, but it may not capture all issues. If the network and name resolution is not setup properly before installation, it is likely the installation will fail; if network or name resolution is malfunctioning, likely the clusterware and/or RAC will have issues. The goal of this note is to provide a list of things to verify regarding the network and name resolution setup for Grid Infrastructure (clusterware) and RAC.

SCOPE

This document is intended for Oracle Clusterware/RAC Database Administrators and Oracle Support engineers.

DETAILS

A. Requirement

Network ping with package size of Network Adapter (NIC) MTU should work on all public and private network.

IP address 127.0.0.1 should only map to localhost and/or localhost.localdomain, not anything else.

127.*.*.* should not be used by any network interface.

Public NIC name must be same on all nodes.

Private NIC name should be same in 11gR2 and must be same for pre-11gR2 on all nodes

Public and private network must not be in link local subnet (169.254.*.*), should be in non-related separate subnet.

MTU should be the same for corresponding network on all nodes.

Network size should be same for corresponding network on all nodes.

As the private network needs to be directly attached, traceroute should work with a packet size of NIC MTU without fragmentation or going through the routing table on all private networks in 1 hop.

Firewall needs to be turned off on the private network.

For 10.1 to 11.1, name resolution should work for the public, private and virtual names.

For 11.2 without Grid Naming Service (aka GNS), name resolution should work for all public, virtual, and SCAN names; and if SCAN is configured in DNS, it shouldn't be in local hosts file.

For 11.2.0.2 and above, multicast group 230.0.1.0 should work on private network; with patch 9974223, both group 230.0.1.0 and 224.0.0.251 are supported. Withpatch 10411721 (fixed in 11.2.0.3), broadcast will be supported. See Multicast/Broadcast section to verify.

OS level bonding is recommended for the private network for pre-11.2.0.2.  Depending on the platform, you may implement bonding, teaming, Etherchannel, IPMP, MultiPrivNIC etc, please consult with your OS vendor for details. Started from 11.2.0.2, Redundant Interconnect and HAIP is introduced to provide native support for multiple private network, refer to note 1210883.1 for details.
 
The commands verifies jumbo frames if it's configured. To know more about jumbo frames, refer to note 341788.1
 

B. Example of what we expect

Example below shows what we expect while validating the network and name resolution setup. As the network setup is slightly different for 11gR2 and 11gR1 or below, we have both case in the below example. The difference between 11gR1 or below and 11gR2 is for 11gR1, we need a public name, VIP name, private hostname, and we rely on the private name to find out the private IP for cluster communication.   For 11gR2, we do not rely on the private name anymore, rather the private network is selected based on the GPnP profile while the clusterware comes up. Assuming a 3-node cluster with  the following node information:

11gR1 or below cluster:

Nodename |Public IP |VIP name |VIP        |Private |private IP1 |private IP2           

         |NIC/MTU   |         |           |Name1   |NIC/MTU     |

---------|----------|---------|-----------|--------|----------------------

rac1     |120.0.0.1 |rac1v    |120.0.0.11 |rac1p   |10.0.0.1    |

         |eth0/1500 |         |           |        |eth1/1500   |

---------|----------|---------|-----------|--------|----------------------

rac2     |120.0.0.2 |rac2v    |120.0.0.12 |rac2p   |10.0.0.2    |

         |eth0/1500 |         |           |        |eth1/1500   |

---------|----------|---------|-----------|--------|----------------------

rac3     |120.0.0.3 |rac3v    |120.0.0.13 |rac3p   |10.0.0.3    |

         |eth0/1500 |         |           |        |eth1/1500   |

---------|----------|---------|-----------|--------|----------------------

11gR2 cluster

Nodename |Public IP |VIP name |VIP        |private IP1 |           

         |NIC/MTU   |         |           |NIC/MTU     |

---------|----------|---------|-----------|------------|----------

rac1     |120.0.0.1 |rac1v    |120.0.0.11 |10.0.0.1    | 

         |eth0/1500 |         |           |eth1/1500   |

---------|----------|---------|-----------|------------|----------

rac2     |120.0.0.2 |rac2v    |120.0.0.12 |10.0.0.2    |

         |eth0/1500 |         |           |eth1/1500   |

---------|----------|---------|-----------|------------|----------

rac3     |120.0.0.3 |rac3v    |120.0.0.13 |10.0.0.3    |

         |eth0/1500 |         |           |eth1/1500   |

---------|----------|---------|-----------|------------|----------

SCAN name |SCAN IP1   |SCAN IP2   |SCAN IP3   

----------|-----------|-----------|--------------------

scancl1   |120.0.0.21 |120.0.0.22 |120.0.0.23 

----------|-----------|-----------|--------------------

Below is what is needed to be verify on each node - please note the example is from a Linux platform:

1. To find out the MTU
/bin/netstat -in
Kernel Interface table
Iface       MTU Met    RX-OK RX-ERR RX-DRP RX-OVR    TX-OK TX-ERR TX-DRP TX-OVR Flg
eth0       1500   0   203273      0      0      0     2727      0      0      0 BMRU

# In above example MTU is set to 1500 for eth0


2. To find out the IP address and subnet, compare Broadcast and Netmask on all nodes
/sbin/ifconfig
eth0      Link encap:Ethernet  HWaddr 00:16:3E:11:11:11
          inet addr:120.0.0.1  Bcast:120.0.0.127  Mask:255.255.255.128
          inet6 addr: fe80::216:3eff:fe11:1111/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:203245 errors:0 dropped:0 overruns:0 frame:0
          TX packets:2681 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:63889908 (60.9 MiB)  TX bytes:319837 (312.3 KiB)
..

In the above example, the IP address for eth0 is 120.0.0.1, broadcast is 120.0.0.127, and net mask is  255.255.255.128, which is subnet of 120.0.0.0 with a maximum of 126 IP addresses.  Refer to Section "Basics of Subnet" for more details.

Note: An active NIC must have both "UP" and "RUNNING" flag.

3. Run all ping commands twice to make sure result is consistent

Below is an example ping output from node1 public IP to node2 public hostname:

PING rac2 (120.0.0.2) from 120.0.0.1 : 1500(1528) bytes of data.
1508 bytes from rac1 (120.0.0.2): icmp_seq=1 ttl=64 time=0.742 ms
1508 bytes from rac1 (120.0.0.2): icmp_seq=2 ttl=64 time=0.415 ms

--- rac2 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1000ms
rtt min/avg/max/mdev = 0.415/0.578/0.742/0.165 ms

3.1 Ping all public nodenames from the local public IP with packet size of MTU
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 rac1
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 rac1
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 rac2
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 rac2
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 rac3
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 rac3


3.2.1 Ping all private IP(s) from all local private IP(s) with packet size of MTU
# applies to 11gR2 example, private name is optional
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  10.0.0.1
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  10.0.0.1
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  10.0.0.2
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  10.0.0.2
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  10.0.0.3
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  10.0.0.3


3.2.2 Ping all private nodename from local private IP with packet size of MTU
# applies to 11gR1 and earlier example
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  rac1p
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  rac1p
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  rac2p
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  rac2p
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  rac3p
/bin/ping -s <MTU> -c 2 -I 10.0.0.1  rac3p


4. Traceroute private network

Example below shows traceroute from node1 private IP to node2 private hostname

traceroute to rac2p (10.0.0.2), 30 hops max, 1472 byte packets
 1  rac2p (10.0.0.2)  0.626 ms  0.567 ms  0.529 ms

MTU size packet traceroute complete in 1 hop without going through the routing table.

Note: traceroute option "-F" may not work on RHEL3/4 OEL4 due to OS bug, refer to note: 752844.1 for details.

4.1 Traceroute all private IP(s) from all local private IP(s) with 
# packet size of MTU - on Linux packet length need to be MTU-28
# applies to 11gR2
/bin/traceroute -s 10.0.0.1  -r -F 10.0.0.1  <MTU-28>
/bin/traceroute -s 10.0.0.1  -r -F 10.0.0.2  <MTU-28>
/bin/traceroute -s 10.0.0.1  -r -F 10.0.0.3  <MTU-28>

4.2 Traceroute all private nodename from local private IP with packet size of MTU
# applies to 11gR1 and earlier example
/bin/traceroute -s 10.0.0.1 -r -F rac1p <MTU-28>
/bin/traceroute -s 10.0.0.1 -r -F rac2p <MTU-28>
/bin/traceroute -s 10.0.0.1 -r -F rac3p <MTU-28>

5. Ping VIP hostname
# Ping of all VIP nodename should resolve to correct IP
# Before the clusterware is installed, ping should be able to resolve VIP nodename but
# should fail as VIP is managed by the clusterware
# After the clusterware is up and running, ping should succeed
/bin/ping -c 2 rac1v
/bin/ping -c 2 rac1v
/bin/ping -c 2 rac2v
/bin/ping -c 2 rac2v
/bin/ping -c 2 rac3v
/bin/ping -c 2 rac3v

6. Ping SCAN name
# applies to 11gR2
# Ping of SCAN name should resolve to correct IP
# Before the clusterware is installed, ping should be able to resolve SCAN name but
# should fail as SCAN VIP is managed by the clusterware
# After the clusterware is up and running, ping should succeed
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 scancl1
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 scancl1
/bin/ping -s <MTU> -c 2 -I 120.0.0.1 scancl1

7. Nslookup VIP hostname and SCAN name
# applies to 11gR2
# To check whether VIP nodename and SCAN name are setup properly in DNS
/usr/bin/nslookup rac1v
/usr/bin/nslookup rac2v
/usr/bin/nslookup rac3v
/usr/bin/nslookup scancl1

8. To check name resolution order
# /etc/nsswitch.conf on Linux, Solaris and hp-ux, /etc/netsvc.conf on AIX
/bin/grep ^hosts /etc/nsswitch.conf
hosts:      dns files

9. To check local hosts file
# If local files is in naming switch setting (nsswitch.conf), to make sure 
# hosts file doesn't have typo or misconfiguration, grep all nodename and IP
# 127.0.0.1 should not map to SCAN name, public, private and VIP hostname

/bin/grep rac1       /etc/hosts
/bin/grep rac2       /etc/hosts
/bin/grep rac3       /etc/hosts
/bin/grep rac1v      /etc/hosts
/bin/grep rac2v      /etc/hosts
/bin/grep rac3v      /etc/hosts
/bin/grep 120.0.0.1  /etc/hosts
/bin/grep 120.0.0.2  /etc/hosts
/bin/grep 120.0.0.3  /etc/hosts
/bin/grep 120.0.0.11 /etc/hosts
/bin/grep 120.0.0.12 /etc/hosts
/bin/grep 120.0.0.13 /etc/hosts

# For 11gR2 example
/bin/grep 10.0.0.1   /etc/hosts
/bin/grep 10.0.0.2   /etc/hosts
/bin/grep 10.0.0.3   /etc/hosts
/bin/grep 10.0.0.11  /etc/hosts
/bin/grep 10.0.0.12  /etc/hosts
/bin/grep 10.0.0.13  /etc/hosts

# For 11gR1 and earlier example
/bin/grep rac1p      /etc/hosts
/bin/grep rac2p      /etc/hosts
/bin/grep rac3p      /etc/hosts
/bin/grep 10.0.0.1   /etc/hosts
/bin/grep 10.0.0.2   /etc/hosts
/bin/grep 10.0.0.3   /etc/hosts


# For 11gR2 example
# If SCAN name is setup in DNS, it should not in local hosts file
/bin/grep scancl1      /etc/hosts
/bin/grep 120.0.0.21 /etc/hosts
/bin/grep 120.0.0.22 /etc/hosts
/bin/grep 120.0.0.23 /etc/hosts

C. Syntax reference

Please refer to below for command syntax on different platform

  Linux:            
    /bin/netstat -in
    /sbin/ifconfig
    /bin/ping -s <MTU> -c 2 -I source_IP nodename
    /bin/traceroute -s source_IP -r -F  nodename-priv <MTU-28>
    /usr/bin/nslookup

  Solaris:        
    /bin/netstat -in
    /usr/sbin/ifconfig -a
    /usr/sbin/ping -i source_IP -s nodename <MTU> 2 
    /usr/sbin/traceroute -s source_IP -r -F nodename-priv <MTU>
    /usr/sbin/nslookup

  HP-UX:            
    /usr/bin/netstat -in
    /usr/sbin/ifconfig NIC
    /usr/sbin/ping -i source_IP nodename <MTU> -n 2
    /usr/contrib/bin/traceroute -s source_IP -r -F nodename-priv <MTU>
    /bin/nslookup
    
  AIX:                
    /bin/netstat -in 
    /usr/sbin/ifconfig -a
    /usr/sbin/ping -S source_IP -s <MTU> -c 2 nodename
    /bin/traceroute -s source_IP -r nodename-priv <MTU>
    /bin/nslookup 
 
  Windows:     
    MTU:  
      Windows XP:          netsh interface ip show interface
      Windows Vista/7:    netsh interface ipv4 show subinterfaces
    ipconfig /all 
    ping -n 2 -l <MTU-28> -f nodename
    tracert
    nslookup

D. Multicast

Started with 11.2.0.2, multicast group 230.0.1.0 should work on private network for bootstrapping. patch 9974223 introduces support for another group 224.0.0.251

Please refer to note 1212703.1 to verify whether multicast is working fine.

As fix for bug 10411721 is included in 11.2.0.3, broadcast is supported for bootstrapping as well as multicast. When 11.2.0.3 Grid Infrastructure starts up, it will try broadcast, multicast group 230.0.1.0 and 224.0.0.251 simultaneously, if anyone succeeds, it will be able to start.


On hp-ux, if 10 Gigabit Ethernet is used as private network adapter, without driver revision B.11.31.1011 or later of the 10GigEthr-02 software bundle, multicast may not work.  Run "swlist 10GigEthr-02" command to identify the current version on your HP server.

 

E. Runtime network issues

OSWatcher or Cluster Health Monitor(IPD/OS) can be deployed to capture runtime network issues.