HP Storage Mirroring Software - Manual (Page 153)

11 - 14
! Failover Processing— Failover Trigger , Failover Method , and Items to Failover
! Miscellaneous Source Machine Options— Manual Intervention , Use .SHR Share Mapping File , and Active Directory
7. Select an IP address or source to be monitored by marking the check box to the left of an item in the Names to Monitor
tree.
8. Highlight one IP address in the Names to Monitor tree and select a Target Adapter that will assume that IP address
during failover. Repeat this process for each IP address that is being monitored.
9. Highlight one IP address in the Names to Monitor tree and specify the Monitor Interval . This setting identifies the
number of seconds between monitor requests sent from the target to the source to determine if the source machine is
online. Repeat this process for each IP address that is being monitored.
10. Highlight one IP address in the Names to Monitor tree and specify the Missed Packets. This setting identifies the
number of monitor replies sent from the source to the target that can be missed before assuming the source machine
has failed. Repeat this process for each IP address that is being monitored.
11. If you are monitoring multiple IP addresses, highlight the machine name and specify the Failover Trigger.
! All Monitored IP Addresses Fail —Failover begins when all monitored IP addresses fail.
! One Monitored IP Address Fails —Failover begins when any of the monitored IP addresses fails.
12. Highlight the machine name and specify the Failover Method option by specifying how the source will impact the target
machine:
! Add Source Identity to Target —The source machine’s identity is added to the target machine’s identity.
! Replace Target Identity with Source —The source machine’s identity replaces the target machine’s identity.
13. If you are monitoring multiple IP addresses, highlight the machine name and specify your Items to Failover options.
! IP Addresses
! Monitored IP Addresses —Failover is only performed on the monitored IP addresses.
! All Addresses —Failover is performed on the monitored and unmonitored IP addresses.
! Server Name —Failover is performed on the server name.
NOTE: The target machine is identified in the Monitor Settings dialog box title bar. All changes made in this dialog
box are for the machine referenced in the title bar. If the wrong target is selected, close the Monitor
Settings dialog box and select a different target from the Failover Control Center window.
NOTE: Current IP Addresses displays the IP address(es) currently assigned to the selected target adapter.
NOTE: If the Replace Target Identity with Source check box is selected and Storage Mirroring is replicating data
from multiple source machines to a single target and one of the sources fail, the Storage Mirroring
failover process automatically notifies the other Storage Mirroring sources of the new identity of the
target so that replication can continue.
NOTE: When multiple IP addresses are monitored, there is the potential for IP address conflicts during
failover. Conflict may occur when the number of IP addresses that trigger failover is less than the
number of IP addresses that are assumed by the target during failover.
For example, suppose a source machine has three IP addresses and two of these are monitored. If a
failover occurs due to a machine failure, there is not an address conflict because all of the IP addresses
have failed and no longer exist. If the failure only occurs on one of the monitored addresses, the other
two IP addresses (monitored and unmonitored) are still affected. The
Items to Failover option forces
these addresses to be assumed by the target as well. These addresses then exist on both the source
and the target. This occurs because the
Items to Failover option allows the selected (monitored) IP
addresses or all of the machine addresses to be failed over, but it does not allow the individual failed
IP address to be failed over. Therefore, when a source machine has fewer IP addresses that trigger
failover than IP addresses that will be assumed, there is a risk of an IP address conflict.
The delay before the failover process begins can be used to determine the cause of the failover, thus
possibly avoiding IP address conflicts.