Developers Guide

Server configuration
30 Dell EMC SC Series with Red Hat Enterprise Linux 7x | CML1071
The new LinuxRHEL-iscsi server object is shown in the LinuxRHEL folder and is ready to have
volumes mapped to it.
3.2.2 iSCSI timeout
In single-path environments, configure the iSCSI daemon to wait and queue I/O for an amount of time
sufficient to accommodate proper failure recovery. For example, a SAN fabric failure or SC Series storage
array failover event can take between 5 and 60 seconds to complete. It is recommended to configure the
iSCSI software initiator to queue I/O for up to 60 seconds before starting to fail I/O requests.
Note: The following configuration settings affect iSCSI connection timeouts.
To adjust this timeout value for a single-path environment, open /etc/iscsi/iscsid.conf and configure the
following parameter to 60 seconds instead of the default value of 10 seconds.
node.conn[0].timeo.noop_out_timeout = 60
Note: The following configuration settings affect iSCSI connection timeouts.
When used in a multipath environments, configure the iSCSI daemon to fail a path in 5 seconds. This
minimizes the latency of waiting for a single path to recover. Since I/O is managed by dm-multipath, it will
automatically resubmit any I/O requests to alternating active iSCSI paths. If all paths are down, dm-multipath
will queue the I/O until a path becomes available. This approach enables an environment to sustain failures at
both the network and storage layers.