Switching between mulitple IPs in an active/active configuration...

[Alex J Early Jr]

We have a SuperDome in a highavailability network configuration. Upon failure of one IP path, we would like to have all network traffic
seemlessly moved to the second IP path. My question is, does MCServiceGuard support this type of configuration? Both IP addresses would be visible to the network.

Alex Early

[Mel Burslan]

serviceguard anything but "seamless"
the definition of serviceguard is "highly available". It is NOT "non-stop"

to attain this functionality, you need to look into HP-APA (auto port aggregation) which adjoins 2 or more network interfaces to give you redundancy or higher data throughput depending on how you configure it.

Not being an APA expert myself, I can suggest, if you have support agreement with HP, call the RC and they can walk you thru the process. It is a little complex but quite straight forward.

HTH

[Mel Burslan]

By the way, I noticed that you placed this question under the databases forum. If you are asking this for database redundancy, I would strongly suggest to talk to your database vendor at the same time. Because, as far as my tiny database knowledge goes, only some specific version/configuration of oracle is able to support such a setup. Beware...

[G. Vrijhoeven]

Hi Alex,

May be i do not understand you correct but if you condigure a fail over interface all traffic will automaticly switch to the stand by interface. If you want both ( multiple) interfaces active at the same time you can take a look at APA. It is a product that needs a license.

more info:

http://docs.hp.com/cgi-bin/fsearch/framedisplay?top=/hpux/onlinedocs/J4240-90016/J4240-90016_top.html&con=/hpux/onlinedocs/J4240-90016/00/00/28-con.html&toc=/hpux/onlinedocs/J4240-90016/00/00/28-toc.html&searchterms=apa&queryid=20040514-121932

Regards,

Gideon

[Victor_5]

As far as I know, it is NO for MC/SG, if I understand you correctly.

[curt larson_1 · Accepted solution]

for service guard this is how it works:

Each node (host system) should have an IP address for each active network interface. This address, known as a stationary IP address, is configured in the node's /etc/rc.config.d/netconf file. A stationary IP address is not transferrable to another node, but may be transferrable to a standby LAN interface card. The stationary IP address is not associated with packages.

In addition to the stationary IP address, you normally assign one or more unique IP addresses to each package. The IP addresses associated with a package are called relocatable IP addresses (also known as package IP addresses or floating IP addresses) because the addresses can actually move from one cluster node to another.

Both stationary and relocatable IP addresses will switch to a standby LAN interface in the event of a LAN card failure. In addition, relocatable addresses (but not stationary addresses) can be taken over by an adoptive node if control of the package is transferred. This means that applications can access the package via its relocatable address without knowing which node the package currently resides on.

IP addresses are configured only on each primary network interface card; standby cards are not configured with an IP address. Multiple IP addresses on the same network card must belong to the same IP subnet.

A local network switch involves the detection of a local network interface failure and a failover to the local backup LAN card. The backup LAN card must not have any IP addresses configured.

So, you question of is active to active failover possible is no. Only active to standby is preformed.

In the case of a local network switch, TCP/IP connections are not lost for Ethernet, but IEEE 802.3 connections will be lost. During the transfer, IP packets will be lost, but TCP (Transmission Control Protocol) will retransmit the packets. In the case of UDP (User Datagram Protocol), the packets will not be retransmitted automatically by the protocol. However, since UDP is an unreliable service, UDP applications should be prepared to handle the case of lost network packets and recover appropriately. Note that a local switchover is supported only between two LANs of the same type. For example, a local switchover between Ethernet and FDDI interfaces is not supported, but a local switchover between 10BT Ethernet and 100BT Ethernet is supported.

[Ted Buis]

The short of it is that ServiceGuard needs an unconfigured LAN that will be reconfigured to replace the failed LAN. That is why it is nice to have 3 or for LAN ports on separate cards for SG. APA should help you if you don't like the SG approach. Full functionality of APA requires a switch with appropriate intelligence like Cisco or Procurve.