Developers Guide

Server configuration
42 Dell EMC SC Series with Red Hat Enterprise Linux 7x | CML1071
In either failover scenario, the SC Series storage may take anywhere from five up to 60 seconds to propagate
these changes through the SAN fabric.
In order to mitigate I/O disruption in multipath scenarios, it is recommended to instruct the HBA code to wait
up to 5 seconds before marking a port as down or failed. This minimizes I/O latency to the I/O or application
stack above it by quickly relocating I/O requests to alternate and still active HBA paths through the SAN
fabric. If all of the paths are down, multipathd will start queuing I/O until one or more paths have recovered.
This allows the SC Series array sufficient time to relocate the WWPN identities of the failed ports to an active
ports and propagate the changes through the SAN fabric. The configuration syntax for a multipath
environment is outlined in the section below.
In SAS-connected environments, paths from both controllers are presented to the connected host
(active/optimized and standby), however only the active/optimized path is used for all active I/O at any one
time. When the active/optimized path becomes unavailable, the SC Series array will dynamically determine
which one of the remaining standby paths will assume the role of the active/optimized path and continue to
stream active I/O to the new active/optimized path.
3.7.1 SC Series volume PortDown timeout
These settings dictate how long a Linux system waits before destroying a connection after being losing its
connectivity with the port. Configure the following in the qla2xxx.conf or lpfc.conf file accordingly.
 For QLogic:
options qla2xxx qlport_down_retry=5
 For Emulex:
options lpfc lpfc_devloss_tmo=5
3.7.2 Verifying parameters
To verify that the configuration changes have taken effect, apply the following commands.
 For QLogic:
# cat /sys/module/qla2xxx/parameters/qlport_down_retry
5
 For Emulex:
# cat /sys/class/scsi_host/hostX/lpfc_devloss_tmo
5
3.8 Single-path environments
During an SC Series path or controller failover event with the SC Series storage operating in legacy port
mode, any ports experiencing failure will trigger their respective WWPN identities to momentarily disappear
from the SAN fabric before relocating these WWPN identities to a reserve port within the same fault domain
on the alternate active controller.
During an SC Series path failover event with SC Series storage operating in virtual port mode, the WWPN
identities of any NPIV ports experiencing failure will be relocated to another active NPIV port within the same
fault domain on the same controller. During a controller failover event, the WWPN identities of any NPIV