this is a good start. but we need more details around:

- best practice for configuration 2x 5424 to act as a primary and secondary switch and the use of LAG or LACP groups (either via copper or fiber ports)

- how to configure VLAN routing for when it comes time to have your iSCSI array replicate it's data to another site (subnet).

- how to configure STP to ensure management uplinks from each switch back to a single core switch

My typical configuration invovles creating the dedicated iSCSI VLAN on the core switch, then setting up an access port uplink on the primary and secondary 5424 on that iSCSI VLAN. I then use the other 3 fiber ports as the LAG/LACP group for the ISL trunk (per Equallogic Guidelines). This essentially gives me a dedicated iSCSI VLAN on the core and the iSCSI switches, and it's routable through the core to be able to route out to external sites for array replication. I can also access the array and other iSCSI devices management from your user networks for day-to-day management, email alerting, etc. etc.

the problem I have is that if the uplink on the primary switch goes doen (or it fails) I lose complete management access to the secondary switch, even though it has it's own dedicated fiber uplink to the core on the iSCSI VLAN.

did you get any more info on your additional questions? my setup will be pretty similar to yours, so I am curious.

Come on Dell, upload a new config, we too have 2 "5424" 's and would like LAG and remote replication setup notes. We use the PS5000XV array. VMware vSphere4

I am in a similar situation with STP configuration and management/production uplinks.

I have two 5424s in a basic vSphere setup where half a switch is dedicated to an iSCSI VLAN and the other half is shared amongst three VLANs handling management/production traffic from the ESX hosts. There are three access ports, one for each VLAN, sending untagged packets to our existing three physical subnets (each with a physical layer 2 switch), respectively. With one VLAN connected to a physical switch, traffic flows ok. As soon as I hook up the 2nd VLAN's access port to its corresponding switch, this access port takes over and nothing gets out from the 1st VLAN. The switch console initially reports the first access port is blocked during the STP learning process but then is forwarding again after 30 secs - however there is no traffic!

I've tried both RSTP and MSTP (one instance per vlan) with the same results, although this is the first time I've really gotten into the guts of STP and may have missed something.