Understanding and Designing Serviceguard Disaster Recovery Architectures

kilometers if Finisar (long haul) GBICs are used for the Inter Switch Links (ISL) between the
Fibre Channel switches. WDM links can also be used for the connection between the Fibre
Channel switches in the two Primary data centers and can provide ISL connections of up to
100 kilometers in length .
There must be at least two TCP/IP networking links, routed geographically differently between
each Primary data center to prevent the “backhoe problem. For example the backhoe problem
can occur when all cables are routed through a single trench and a tractor on a construction
job severs all cables and disables all communications between the data centers. Only a single
network link can be routed from each Primary data center to the Arbitrator data center.
However, to survive the loss of the network link between a Primary data center and the
Arbitrator data center, the network routing must be configured so that a Primary data center
can also reach the Arbitrator via a route which passes through the other Primary data center.
There must be at least two Fibre Channel Data links, routed geographically differently between
each data center. In three data center configurations, no Fibre Channel Data links are required
for the Arbitrator data center.
Redundant Fibre Channel switches are required in each data center, unless the switch offers
built-in redundancy.
Refer to the SWD Streams documents for supported Fibre Channel switches An Extended
Fabric license may be required if the inter-switch links (ISL) between the switches is greater
than 10 kilometers. For optimum data replication performance, it is suggested to tune the
buffer credits properly for the ISL used for data replication between the data centers.
If CVM or CFS is being used and all data replication links are lost between the data centers,
but the network links remain functional, it is likely that all mirror copies that the CVM master
cannot contact will be detached from the disk group. This means that applications running on
nodes which are not in the same data center as the CVM Master node may hang, as their
local mirror copies are detached and they cannot reach the remote mirror copies. SLVM
handles this scenario differently, because it will arbitrate (during which time all writes to the
shared volumes will temporarily hang) and allow only one of the nodes to continue writing to
the shared volumes (the writes on the other node will continue to hang until the data replication
links are re-established).
Common WDM Links for both TCP/IP Networking and Fibre Channel Data
The maximum distance supported between the data centers for DWDM and CWDM
configurations is 100 kilometers.
Both the TCP/IP networking and Fibre Channel Data can go through the same WDM box.
WDM hardware is typically designed to be fault tolerant, therefore it is acceptable to use only
one WDM box (in each data center) for the links between each data center. However, for the
highest availability, HP recommends you to have redundant WDM boxes (in each data center)
used for the links between each data center. If you are using a single WDM box for the links
between each data center, you must ensure that no Single Points of Failure (SPOFs) exist for
that WDM box, and the redundant standby fiber link feature of the WDM box must be
configured. If the WDM box supports multiple active WDM links, that feature can be used
instead of the redundant standby feature.
At least two fiber
4
optic links are required between each Primary data center, each fiber link
routed geographically differently to prevent the “backhoe problem. Only one fiber link routed
from each Primary data center to the Arbitrator data center, however in order to survive the
loss of a link between a Primary data center and the Arbitrator data center, the network routing
must be configured so that a Primary data center can also reach the nodes in the Arbitrator
data center via a route passing through the other Primary data center.
60 Extended Distance Cluster Configurations