Oracle VM 2.2 Nic bonding

Hi everybody,
I want to set NIC bonding on my VM 2.2 server. How does NIC bonding work? Also would I have to do any setup on my switch? Thank you to anyone who can help me understand.

Sean314 wrote:
I want to set NIC bonding on my VM 2.2 server. How does NIC bonding work? Also would I have to do any setup on my switch? Thank you to anyone who can help me understand.There is a great document on the Oracle Wiki for this: [ Oracle VM Server Configuration- bonded and trunked network interfaces|http://wiki.oracle.com/page/Oracle+VM+Server+Configuration-+bonded+and+trunked+network+interfaces].

Similar Messages

GUIDE: Oracle VM VirtualBox and Oracle Linux NIC bonding

Author: Dude
Date: 01-Jun-2013
Version: B
Last updates:
23-Jul-2013, changes required due to forum upgrade
The following is provided in the hope that it might be useful. If you wish to use it for the purpose of reference, presentation or education, please include the link or information to this forum posting. Thank you!
What is Linux network interface bonding? Simply put, the Linux system provides a kernel driver, which can create a logical network device (bond) using 2 or more Network Interface Controllers (NIC) to accomplish link-aggregation and adapter fault-tolerance. For more information, please see the Deployment Guide at https://linux.oracle.com/documentation/
The channel bonding interface supports several modes of operation. In a typical active-backup set up for fault-tolerance, the system performs MII-based health monitoring of the network interface slaves (eth0, eth1) and assigns them according to bond options and interface availability. In active-backup mode (mode 1), only the current active network device is seen externally.
Linux bonding for fault-tolerance under Oracle VM VirtualBox does not seem to function as advertised. Searching the Internet for more information shows that other people have tried to set up Linux bonding using various virtualization products, virtual host network adapter and bonding options, but without success.
It appears the management of Media Access Control addresses (MAC) interferes with internal routing methods between the host and the virtual machine guest. Using the modinfo bonding command (fig. 1) shows a listing of available parameters for the bonding kernel module. The fail_over_mac parameter addresses the Mac address issue in VirtualBox.
Fig.1
# modinfo bonding
filename:       /lib/modules/2.6.39-400.17.1.el6uek.x86_64
                /kernel/drivers/net/bonding/bonding.ko
parm:           fail_over_mac:For active-backup,
                do not set all slaves to the same MAC;
                0 for none (default), 1 for active, 2 for follow (charp)
Configuration Example
Virtualization Product: Oracle VM VirtualBox 4.2.10
VM Guest OS: Oracle Linux 6.4 (x64)
VM adapter settings: Bridged Adapter, Paravirtualized Network (virtio-net)
Adapter1 (eth0): 080027436617, en0, Ethernet 1
Adapter2 (eth1): 08002708F36A, en0, Ethernet 1
Note: If you wish to reset the network adapter names and MAC addresses used by the Oracle Linux guest system, erase the content of /etc/udev/rules.d/70-persistent-net.rules and restart the virtual machine.
1. Create or modify the following configuration files:
/etc/modprobe.d/bonding-eth0_eth1.conf
alias bond0 bonding
/etc/sysconfig/network-scripts/ifcfg-bond0
DEVICE=bond0
IPADDR=10.0.2.12
NETMASK=255.0.0.0
GATEWAY=10.0.0.138
DNS1=10.0.0.138
DNS2=8.8.8.8
ONBOOT=yes
BOOTPROTO=none
USERCTL=no
BONDING_OPTS="mode=1 miimon=100 fail_over_mac=1"
/etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0
BOOTPROTO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
USERCTL=no
/etc/sysconfig/network-scripts/ifcfg-eth1
DEVICE=eth1
BOOTPROTO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
USERCTL=no
2. Restart the system or network services (root):
Fig. 2
# service network restart
Configuration Test
Enter the following commands as user root to verify proper operation of NIC bonding:
Fig. 3
# tail /var/log/messages
bond0: setting mode to active-backup (1).
bond0: Setting MII monitoring interval to 100.
bond0: Setting fail_over_mac to active (1).
bond0: Adding slave eth0.
bond0: making interface eth0 the new active one.
bond0: first active interface up!
bond0: enslaving eth0 as an active interface with an up link.
bond0: Adding slave eth1.
bond0: enslaving eth1 as a backup interface with an up link.
Fig. 4
# ifconfig
bond0     Link encap:Ethernet HWaddr 08:00:27:43:66:17
          inet addr:10.0.2.12 Bcast:10.255.255.255 Mask:255.0.0.0
          inet6 addr: fe80::a00:27ff:fe43:6617/64 Scope:Link
          UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
eth0      Link encap:Ethernet HWaddr 08:00:27:43:66:17
          UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
eth1      Link encap:Ethernet HWaddr 08:00:27:08:F3:6A
          UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
Fig. 5
# netstat -rn
Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window irtt Iface
0.0.0.0         10.0.0.138      0.0.0.0         UG        0 0          0 bond0
10.0.0.0        0.0.0.0         255.0.0.0       U         0 0          0 bond0
169.254.0.0     0.0.0.0         255.255.0.0     U         0 0          0 bond0
Fig. 6
# ifconfig eth0 down
# tail /var/log/messages
bond0: link status definitely down for interface eth0, disabling it
bond0: making interface eth1 the new active one.
# ifconfig
bond0     Link encap:Ethernet HWaddr 08:00:27:08:F3:6A
          inet addr:10.0.2.12 Bcast:10.255.255.255 Mask:255.0.0.0
          inet6 addr: fe80::a00:27ff:fe43:6617/64 Scope:Link
          UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
eth1      Link encap:Ethernet HWaddr 08:00:27:08:F3:6A
          UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
Fig. 7
# ifconfig eth0 up
# ifconfig eth1 down
# tail /var/log/messages
bond0: link status definitely up for interface eth0, 4294967295 Mbps full duplex.
bond0: link status definitely down for interface eth1, disabling it
bond0: making interface eth0 the new active one.
# ifconfig
bond0     Link encap:Ethernet HWaddr 08:00:27:43:66:17
          inet addr:10.0.2.12 Bcast:10.255.255.255 Mask:255.0.0.0
          inet6 addr: fe80::a00:27ff:fe43:6617/64 Scope:Link
          UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
eth0      Link encap:Ethernet HWaddr 08:00:27:43:66:17
          UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
During the above test, the client network connection from host 10.0.0.1 (ssh) was maintained without any noticeable interruption. As it can be seen from the syslog and ifconfig command output, the bond0 interface was automatically assigned either the MAC address of eth0 or eth1 and the fail-over worked as advertised.
Good luck!

Please see these docs/links.
Using Oracle VM with Oracle E-Business Suite Release 11i or Release 12 [ID 465915.1]
Certified Software on Oracle VM [ID 464754.1]
Oracle VM Templates -- Oracle E-Business Suite 12.1.3 (32-bit), Oracle E-Business Suite 12.1.1 (64-bit and 32-bit) and Oracle E-Business Suite Release 12.X Sparse Middle Tier (32-bit)
http://www.oracle.com/technetwork/server-storage/vm/ebs-093153.html
E-Business Suite 12.1.3 Templates for Oracle VM Now Available
http://blogs.oracle.com/stevenChan/2010/11/ebs1213_oraclevm.html
Live Migration of EBS Services Using Oracle VM
http://blogs.oracle.com/stevenChan/2010/01/ebs_live_migration_ovm.html
Using Oracle VM with Oracle E-Business Suite Virtualization Kit
http://blogs.oracle.com/stevenChan/2009/12/oracle_vm_ebs_virtualization_kit.html
New Oracle® VM Templates for Oracle Applications Now Available
http://www.oracle.com/us/corporate/press/068341
Very Cool! New Oracle VM Templates for Oracle E-Business Suite
http://blogs.oracle.com/virtualization/2010/11/very_cool_new_oracle_vm_templa.html
Thanks,
Hussein

Nic Bonding in OVM Server

Hi
I am setting up a OVM environment where I want to implement NIC bonding. I want to bond eth1 and eth3, the IP address are showing against xenbr1 and xenbr3. Please let me know the steps of doing this.
Thanks

itsakc wrote:
I am setting up a OVM environment where I want to implement NIC bonding. I want to bond eth1 and eth3, the IP address are showing against xenbr1 and xenbr3. Please let me know the steps of doing this.The Underground Oracle VM Manual has an entire chapter on this: http://itnewscast.com/chapter-7-oracle-vm-networking-8021q

NIC Bonding not behaving as it should

Hi Folks, am trying to setup NIC bonding on my VM (OEL5u6) having two virtualised NICs, i have done the setup which is quite simple however my active-backup mode is not working as expected.
My Bonded interface
[root@Gateway network-scripts]# cat ifcfg-bond0
DEVICE=bond0
IPADDR=192.169.25.2
NETMASK=255.255.255.248
ONBOOT=yes
BOOTPROTO=none
USERCTL=no
BONDING_OPTS='mode=active-backup miimon=1000'
TYPE=Ethernet
individual interfaces
[root@Gateway network-scripts]# cat ifcfg-eth2
DEVICE=eth2
HWADDR=08:00:27:1e:57:cf
MASTER=bond0
SLAVE=yes
ONBOOT=yes
BOOTPROTO=none
USERCTL=no
[root@Gateway network-scripts]# cat ifcfg-eth4
DEVICE=eth4
HWADDR=08:00:27:46:69:69
MASTER=bond0
SLAVE=yes
ONBOOT=yes
BOOTPROTO=none
USERCTL=no
Bonding status (looks good)
[root@Gateway network-scripts]# cat /proc/net/bonding/bond0
Ethernet Channel Bonding Driver: v3.5.0 (November 4, 2008)
Bonding Mode: fault-tolerance (active-backup)
Primary Slave: None
Currently Active Slave: eth2
MII Status: up
MII Polling Interval (ms): 1000
Up Delay (ms): 0
Down Delay (ms): 0
Slave Interface: eth2
MII Status: up
Link Failure Count: 0
Permanent HW addr: 08:00:27:1e:57:cf
Slave Interface: eth4
MII Status: up
Link Failure Count: 0
Permanent HW addr: 08:00:27:46:69:69
I am doing a continous ping from another host in the same n/w to this bonded interface IP to check if the ping streak breaks when i test this setup
Testing -
[root@Gateway network-scripts]# ifenslave -c bond0 eth4
and that's it my SSH connection is gone, and ping is broken with "Request timed out"
when i check the bond status now -
Bonding Mode: fault-tolerance (active-backup)
Primary Slave: None
Currently Active Slave: eth4
MII Status: up
MII Polling Interval (ms): 1000
Up Delay (ms): 0
Down Delay (ms): 0
Slave Interface: eth2
MII Status: up
Link Failure Count: 0
Permanent HW addr: 08:00:27:1e:57:cf
Slave Interface: eth4
MII Status: up
Link Failure Count: 0
Permanent HW addr: 08:00:27:46:69:69
am afraid there's not much help in the message log to debug this.
Any views what's wrong and how to resolve? appreciate ur replies. Thanks
Regards Amit

Hi,
I'm having the same problem, when i disconnect the cable from the eth0 (active slave) stop receiving respond
I have tried bridged conf and interal net.
I'm using VirtualBox 4.1.16
This is my conf:
[root@integrador ~]# cat /etc/sysconfig/network-scripts/ifcfg-bond0
DEVICE=bond0
IPADDR=10.10.0.2
NETWORK=10.10.0.0
NETMASK=255.255.255.0
BONDING_OPTS="mode=1 miimon=100"
BOOTPROTO=none
ONBOOT=yes
[root@integrador ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0
ONBOOT=yes
BOOTPROTO=none
MASTER=bond0
SLAVE=yes
USERCTL=no
[root@integrador ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth1
DEVICE=eth1
ONBOOT=yes
BOOTPROTO=none
MASTER=bond0
SLAVE=yes
USERCTL=no
[root@integrador ~]# cat /etc/modprobe.d/bonding.conf
alias bond0 bonding
[root@integrador ~]# ifconfig -a
bond0 Link encap:Ethernet HWaddr 08:00:27:5C:C5:BA
inet addr:10.10.0.2 Bcast:10.10.0.255 Mask:255.255.255.0
inet6 addr: fe80::a00:27ff:fe5c:c5ba/64 Scope:Link
UP BROADCAST RUNNING MASTER MULTICAST MTU:1500 Metric:1
RX packets:402 errors:0 dropped:0 overruns:0 frame:0
TX packets:92 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:39487 (38.5 KiB) TX bytes:11045 (10.7 KiB)
eth0 Link encap:Ethernet HWaddr 08:00:27:5C:C5:BA
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:211 errors:0 dropped:0 overruns:0 frame:0
TX packets:89 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:23061 (22.5 KiB) TX bytes:10863 (10.6 KiB)
eth1 Link encap:Ethernet HWaddr 08:00:27:5C:C5:BA
UP BROADCAST RUNNING SLAVE MULTICAST MTU:1500 Metric:1
RX packets:191 errors:0 dropped:0 overruns:0 frame:0
TX packets:3 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:16426 (16.0 KiB) TX bytes:182 (182.0 b)
eth2 Link encap:Ethernet HWaddr 08:00:27:DF:11:A2
inet addr:192.168.57.1 Bcast:192.168.57.3 Mask:255.255.255.252
inet6 addr: fe80::a00:27ff:fedf:11a2/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:1142 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 b) TX bytes:119508 (116.7 KiB)
eth3 Link encap:Ethernet HWaddr 08:00:27:D1:41:9A
inet addr:192.168.1.50 Bcast:192.168.1.255 Mask:255.255.255.0
inet6 addr: fe80::a00:27ff:fed1:419a/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:825 errors:0 dropped:0 overruns:0 frame:0
TX packets:625 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:563221 (550.0 KiB) TX bytes:85914 (83.9 KiB)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:25 errors:0 dropped:0 overruns:0 frame:0
TX packets:25 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:2690 (2.6 KiB) TX bytes:2690 (2.6 KiB)
[root@integrador ~]# cat /proc/net/bonding/bond0
Ethernet Channel Bonding Driver: v3.6.0 (September 26, 2009)
Bonding Mode: fault-tolerance (active-backup)
Primary Slave: None
Currently Active Slave: eth0
MII Status: up
MII Polling Interval (ms): 100
Up Delay (ms): 0
Down Delay (ms): 0
Slave Interface: eth0
MII Status: up
Speed: 1000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: 08:00:27:5c:c5:ba
Slave queue ID: 0
Slave Interface: eth1
MII Status: up
Speed: 1000 Mbps
Duplex: full
Link Failure Count: 0
Permanent HW addr: 08:00:27:fc:6b:f6
Slave queue ID: 0
[root@integrador ~]# lsmod | grep -i bond
bonding 109558 0
ipv6 264641 39 cnic,bonding,ip6t_REJECT,nf_conntrack_ipv6,nf_defrag_ipv6
[root@integrador ~]# uname -a
Linux integrador 2.6.32-220.el6.i686 #1 SMP Tue Dec 6 16:15:40 GMT 2011 i686 i686 i386 GNU/Linux
Sorry for my english!

CSACS-3415 ACS 5.4 NIC Bonding / Teaming possible ?

Hi Team,
I know, this topic has been answered for the "old" 11x Appliances: not possible.
Does the new UCS hardware change anything ?
Can we bundle 2 NICs somehow to get interface redundancy ?
If still not possible to configure that in ACS 5 itself:
Can it enentually be done on the "hardware" level
within the appliance firmware (UCS BIOS) ?
Frank
(RHEL would provide NIC bonding,,, unfortunately its not accessable from ACS5 CLI)

Yes it does. ACS 5.5 with the Cisco SNS-3415, Cisco SNS-3495, virtual machine, or CSACS-1121 platform allows you to use up to four network interfaces: Ethernet 0, Ethernet 1, Ethernet 2, and Ethernet 3.
NIC Bonding
http://www.cisco.com/c/en/us/td/docs/net_mgmt/cisco_secure_access_control_system/5-5/installation/guide/csacs_book/csacs_hw_ins.html#pgfId-1191791
Creating interface bonding
http://www.cisco.com/c/en/us/td/docs/net_mgmt/cisco_secure_access_control_system/5-5/installation/guide/csacs_book/csacs_hw_ins.html#pgfId-1197533
Regards,
Jatin Katyal
*Do rate helpful posts*

Switching from single NIC to NIC bonding in "active" RAC

Hi
I have a 4 node RAC cluster on RHEL 4, running 10.2.0.3.0 in "semi-production".
"Everything" is multiplexed, except the NICs. The servers have 2 NIC ports one for public and one for interconnect.
If we add another NIC with 2 ports, is it possible to change the NIC configuration from single port to NIC bonding, without reinstalling CRS?
I can get a maintenance window, so this doesn't have to be done on the fly, I just need to know if its do-able.
I have checked some notes on MetaLink and for me it looks like this should be possible.
E.g. Note:436555.1 and Note:283684.1
Regards
Niklas

Nicke, Its very well doable and supported approach. You can use oifcfg to reconfigure the interfaces and there is no need to reinstall the CRS.
-Harish Kalra

Oracle VM 3.1.1 Manager: Configuraing NIC Bonding?

Each of the two OVM 3.1.1 servers in the OVMM 3.1.1 server pool has three Gigabit Ethernet NIC's:
<ol>
<li>One for the Management + VM + Heartbeat Networks (eth0) and
<li>The other two (eth1 + eth2) should be bonded for redundancy and performance and connected to the Live Migration + Storage Networks.
</ol>
However, OVMM 3.1.1 doesn't seem to have the ability to configure the bonding for NIC's. Am I overlooking something?
Eric Pretorious
Truckee, CA

Refer to the Oracle VM User's Guide for Release 3.1.1:
<li>Chapter 5.4, Network Bonding discusses bonding Management interfaces.
<li>Chapter 5.8, Managing Bonded Interfaces discusses bonding non-Management interfaces.
Eric Pretorious
Truckee, CA
REFERENCE
http://docs.oracle.com/cd/E27300_01/E27309/html/vmusg-network-bonding.html
http://docs.oracle.com/cd/E27300_01/E27309/html/vmusg-network-bondintfce.html

Oracle AVDF Database Firewall - NIC's

Hi all,
We are using Oracle Audit Vault & Database Firewall 12.1.0.Currently we have only two NICs available for Database Firewall. What can we do with this number of NICs? Do we need three for DAM mode ? Not according Oracle Docs, but from your experience please, we have already read the docs...
Thank You.

Hi,
I had to login with root and do the next:
cd /
find * -mtime -1 | more
this show the file modified on the last hour .... in my case does not work because I had a problem with conditions on alerts.

ISE 1.2 SNS-3415 NIC Bonding / Teaming

Hello,
I have installed the SNS-3415 with ISE 1.2 and i'm trying to setup redundnacy (Team) nic modes for the authentication requests and not for management purpose.
The tests showed that when the one interface was unpluged everything was lost and nobody from our internal users was able to authenticated by the ISE node.
In contrast when i was unpluged the " second interface " (probably the inactive ) nothing was happened which shows that is a useless interface
My purpose is to connect it to my twins core switches and have a full high availability deployment.
- I have search enough on the WEB but i didn't found any clear and precisely document of saying how this could be achieved.
http://www.cisco.com/c/en/us/td/docs/net_mgmt/cisco_secure_access_control_system/5-4/installation/guide/csacs_book/csacs_hw_ins_ucs.html#wp1185589
Themis

ISE 1.2 does not support NIC teaming. Especially on appliances. There is a workaround for VM using the ESXi host to team the NICs so that it is transparent to the VM.
Please Rate Helpful posts and mark this question as answered if, in fact, this does answer your question. Otherwise, feel free to post follow-up questions.
Charles Moreton

New rc.conf networking and nic bonding

I understand that net-tools is deprecated (actually a good idea),
but i have the feeling that someone forget to think about bonds.
Netcfg should support it (optionally), but i am not able to get it working (lack of documentation).
It is possible to make the new method working:
in rc.conf:
interface=bond0
address=
then in rc.local
echo +eth0 > /sys/class/net/bond0/bonding/slaves
echo +eth1 > /sys/class/net/bond0/bonding/slaves
ip link set dev bond0 up
but i don't think that is *the* way to go
is there a better way to do it ?
maybe a feature request ?

geometric wrote:I think that optimal decision would be both ways of configuring network.
Why developers wouldn't do that? the old way is more flexible anyway
"Optimal" from who's point of view?
Have you read http://www.archlinux.org/news/deprecation-of-net-tools/ ?

NIC Teaming/Bonding Support by Oracle or NOT

hi:
I have customers using RedHat Linux ES3 Update 6 with NIC Bonding/Teaming on the Oracle 10G RAC's private/heartbeat LAN, I would like to hear from you all or Oracle Support whether SUCH CONFIGURATION IS SUPPORTED BY ORACLE OR NOT.
Thanks.
Bennie

hi:
I have customers using RedHat Linux ES3 Update 6 with NIC Bonding/Teaming on the Oracle 10G RAC's private/heartbeat LAN, I would like to hear from you all or Oracle Support whether SUCH CONFIGURATION IS SUPPORTED BY ORACLE OR NOT.
Thanks.
Bennie

Channel Bonding? Oracle RAC with multiple redundant network

We are planning on setting up on Oracle RAC 10g on Linux.
We have redundant switches, so I would like to set up the network to have some version of network redundancy.
Under windows, it is called 'teaming'. Linux is 'channel-bonding'
I just want to make sure if I setup the channel-bonded interfaces, it will be supported with RAC. Has anyone done this before??
-Andy

we are running our RAC on NIC bonding. Although we need to do some more testing to be real sure as to the functionality of the bond.
Metalink Note:298891.1 would be a good place to start.
hth,
-S

NIC teaming/bonding for interconnection

Hi to all,
We are planning for a two node Oracle RAC implementation on Windows Server and we have some doubts about interconnect. Because redundancy is not as important, and because we are limited by money Gigabit switch will be used for interconnection and for public network.
Can we get in bandwidth (load balancing) if we use NIC Bonding on one switch. Which is best?
- To buy servers with two single 1Gb port NIC’s and bond this NIC’s for interconnection and one single port NIC for public network or
- One dual port NIC (bond this NIC’s) for interconnects and one single NIC for public network?

Nikoline wrote:
We are planning for a two node Oracle RAC implementation on Windows Server and we have some doubts about interconnect. Because redundancy is not as important, and because we are limited by money Gigabit switch will be used for interconnection and for public network.Keep in mind that the h/w architecture used, directly determines how robust RAC is, how well it performs, and how it meets redundancy, high availability and scalability requirements.
The wrong h/w choices can and will hurt all of these factors.
Can we get in bandwidth (load balancing) if we use NIC Bonding on one switch. No. Bonding usually does not address load balancing. You need to check exactly what the driver stack used for bonding supports.
The primary reason for bonding is redundancy. Not load balancing/load sharing.
Which is best?
- To buy servers with two single 1Gb port NIC’s and bond this NIC’s for interconnection and one single port NIC for public network or
- One dual port NIC (bond this NIC’s) for interconnects and one single NIC for public network?A server with dual 1Gb NICs will be bonded into a single bonded NIC. And that is what then must be used - the logical bonded NIC and not the individual NICs.
This means that 2 1Gb ports provide you with a single bonded port. And you need 2 ports for RAC - public and interconnect.
So you either need 4 physical ports in total to create 2 bonded ports, or 3 ports in total for 1 bonded port and 1 unbonded port.
And what about your cluster's shared storage?
This, together with the Interconnect, determines the robustness and performance of RAC. 1Gb Interconnect is already pretty much a failure in providing a proper Interconnect infrastructure for RAC.
Keep in mind that the Interconnect is used to share memory across cluster nodes (cache fusion). The purpose is to speed up I/O - making it faster for a cluster node to get data blocks from another node's cache instead of having to hit spinning rust to read that data.
Old style fibre channel technology for shared storage is dual 2Gb fibre channels. Which will be faster than your 1Gb Interconnect. How does it make sense to use an Interconnect that is slower (less bandwidth and more latency) than the actual I/O fabric layer?
Would you configure h/w for a Windows server (be that a database, web or mail server) where the logical I/O from the server's memory buffer cache for local disks is slower than actually reading that data off local disk?
Then why do this for RAC?
Last comment - over 90% of the 500 fastest and biggest clusters on this planet run Linux. 1% runs Windows. Windows is usually a poor choice when it comes to clustering. Even Oracle does not provide their Exadata Database Machine product (fastest RAC cluster in the world) on Windows - only on Linux and Solaris (and the latter only because Oracle now owns Solaris too).

NIC ROUTER requirement for 10g RAC database

I need to know what hardware to order in setting up an 10g RAC database without a single-point-of-failure. My question centers on the networks. If each server has one NIC for the public interface and one NIC for the private interconnect, aren't the routers these NICs are attached to a single point of failure? If each server has two NICs bonded together for the public interface and two NICs bonded together for the private interconnect, does each NIC attach to a different router?

sayantan chakraborty wrote:
is RDMA and infiband ar same??Infiniband is a switched fabric layer for high speed communication. RDMA is a protocol for "+remote direct memory access+".
Infiniband supports the Internet Protocol suite over Infiniband, or IPoIB. It also supports RDMA over Infiniband.
so do we need only infiband SDR / QDR switch to set infiband OR what else??You need an Infiniband switch - Cisco and Voltaire are two companies that supply this type of hardware. The latter is used by Oracle in its Exadata database machines. Unsure about Cisco's commitment to Infiniband switches as they have discontinued their Infiniband switches with integrated fibre channel gateways. A sore point with us after we bought into this technology about 2 years ago, in part on their very own recommendations - and now own very expensive, somewhat buggy and totally unsupported hardware. Seems like the new Cisco California servers are GigE and not Infiniband, which perhaps explain this (flawed) decision of theirs...
For each cluster node, you need an Infiniband PCI card (typically has 2 ports). If you are going to use bonding, you will need a pair of Infiniband cables per server. Also remember to get spares - both PCI cards and cables.
If you want redundancy at the switch, you need to get yourself 2 Infiniband switches.
But if you're planning to invest this into a RAC cluster, then surely you should also invest money in RAC licensing and support - and with that gain access to very useful documentation like the RAC Starter Kit and so on.

Oracle 9i rac interconnect

Hai,
I need clarification in oracle 9i rac internetconnect. Recently we have done installation of oracle 9i rac (9.2.0.4) on aix 5.3 with hacmp cluster for 2 nodes. We are not able to mount the second instance of the database after starting the first instance.
The problem was solved after adding cluster_interconnect parameter in the init.ora file. In all other servers, it is not given and oracle is selecting the interconnect automatically. For this new server alone this problem. I want to know from whether oracle is taking this interconnect if we dont mention the cluster_interconnect paramter. Even i have verified /etc/hosts file and entries are available in it. Requeting your help in this regard.

Where the IP information is stored and managed from depends on your implementation such as are you using a third party clusterware. If you are using Oracle clusterware its stored in the OCR.
You can check this using the following query...
SELECT * FROM GV$CLUSTER_INTERCONNECTS;
The source column will tell you where the interconnect information was obtained from.
BTW, if you are using the CLUSTER_INTERCONNECTS parameter you may loose some of the HA features here is something from the 10gR2 documentation..
" Failover and Failback and CLUSTER_INTERCONNECTS
Some operating systems support run-time failover and failback. However, if you use the CLUSTER_INTERCONNECTS initialization parameter, then failover and failback are disabled."
Question, have you tried other options such as NIC pairing and bonding to have dual interconnects instead of using the CLUSTER_INTERCONNECTS parameter.
Please check Metalink Note # 298891.1 talks about configuring NIC bonding on Linux there are similar options for other operating systems.
answered by Murali Vallath refer the link: this will be very useful to you.http://kr.forums.oracle.com/forums/thread.jspa?threadID=625931
hope, this will helps you.

Oracle VM 2.2 Nic bonding

Similar Messages

Maybe you are looking for