I have a pair of Dell 6224's with stacking modules and a PS6000E.

My intention is to follow the various guides so that hopefully I end up with a single, yet redundant, switch stack with the default VLAN and management IP on our main LAN, with most of the ports assigned to a dedicated VLAN for the iSCSI traffic with the PS6000 connected to these ports across both switches etc.

Where I'm getting a little bogged down is the stacking.

I have my first switch configured with the basics, which is simply its management IP and NTP server settings.

Do I need to run the setup wizard and configure the second switch with its own unique management IP or can I just connect it via the stacking cables, switch it on, and it'll sort itself out?

What happens if the master fails and other takes over, does it take on the management IP of the original master?

Sorry, just not finding the manuals too helpful on the "nitty gritty" of stacking so any clarification would be much appreciated.

When you stack these switches, they will act as a single switch. The ports on the second switch will start with 2 (as in "module" 2) and it will appear as just another set of ports. VLAN interfaces and IP addresses will belong to the whole stack, not any particular switch. Therefore to manage the stack you can telnet to any IP address you have assigned to it. If you have ever worked with a chassis-based switch like a Cisco 4500 or 6500, each switch in a stack behaves like a module in a chassis-based switch.

When the master fails, the standby switch will reboot first (takes about 30-60 seconds) to come up as the master of the stack at which point all your VLAN interfaces will be available. Of course, any physical interfaces that are on the original master will be unavailable if the switch is down. If you are trying to provide any kind of external redundancy to hosts or other equipment via multiple network links, you'll want to have them spread between each physical switch in the stack.

Thanks Mbell98 -- beat me to it. We stacked a couple of 6248's here in the DellTechCenter lab for our EqualLogic groups. Its exactly as mbell98 stated above, once you plug the cable in, the 2nd switch just becomes a part of the primary.

Thanks both - connected the second one up this morning and it appears to have been picked up as hoped/expected.

If anyone's reading who's good on the Equallogic requirements, a bit of a sanity check would be nice.

My config below is intended to give me a stack of two switches, with each split in two with ports 13-24 being on an isolated VLAN which will be used solely for the iSCSI SAN with ports 1-12 on the default VLAN (in reality I only intend connecting one port per switch to the main LAN for management access, it won't be used for production traffic).

I've truncated the output so it will fit, in reality it includes ports 13-24 on members 1 and 2:

console#show running-config
!Current Configuration:
!System Description "Dell 24 Port Gigabit Ethernet, 2.2.0.3, VxWorks5.5.1"
!System Software Version 2.2.0.3
!
configure
vlan database
vlan 2
exit
sntp unicast client enable
sntp server 10.x.x.x
stack
member 1 1
member 2 1
exit
ip address 10.x.x.x 255.255.0.0
ip default-gateway 10.x.x.x
interface vlan 2
name "equallogic"
exit
username "admin" password somestring level 15 encrypted
flowcontrol


!
interface ethernet 1/g13
spanning-tree portfast
switchport access vlan 2
exit
!
interface ethernet 1/g24
spanning-tree portfast
switchport access vlan 2
exit
!
interface ethernet 2/g13
spanning-tree portfast


switchport access vlan 2
exit
!
interface ethernet 2/g24
spanning-tree portfast
switchport access vlan 2
exit
snmp-server community public rw
exit

Hey Hutchingsp,

These links might be of help:

Network Design Considerations for EqualLogic SANs - Part 1
http://en.community.dell.com/blogs/dell_tech_center/archive/2009/08/21/network-design-considerations-for-equallogic-sans-part-1.aspx

Network Design Considerations for EqualLogic SANs - Part 2
http://en.community.dell.com/blogs/dell_tech_center/archive/2009/08/26/network-design-considerations-for-equallogic-sans-part-2.aspx

Check out this page (the same commands applies to the 62xx series)
http://www.delltechcenter.com/page/Configuring+a+PowerConnect+5424+or+5448+Switch+for+use+with+an+iSCSI+storage+system#clearStyles

Also, Check out Matt Ellerback's post on what to do with your EqualLogic SAN,
http://michaelellerbeck.com/2009/11/23/so-you-bought-an-equallogic-san-now-what-part-one/

Which has a link to this key document - Dell EqualLogic Configuration Guide- that also has the exact commands for your switch (in the Appendix):
http://go2.wordpress.com/?id=725X1342&site=mellerbeck.wordpress.com&url=http%3A%2F%2Fsites.google.com%2Fsite%2Fmellerbeck%2FHome%2FDell_EqualLogic_Configuration_Guide.pdf%3Fattredirects%3D0%26d%3D1\

Thanks Jeff - looks along the lines of what I've done (I followed the Equallogic guide that has a section on the 6200 series in it).

I guess my concern was that it tells you to enter the commands but unless you know how to interpret a printout of the configuration it's hard to know 100% that everything is in place hence I thought I'd post the "show running-config" output.

Hi,

I also have a new PS6000 and 2 PowerConnect 6224 switches (with stacking modules)

How are you connecting the ports on the PS6000? I believe all 4 can be used for iSCSI. So I assume that means 2 to each switch?? The guide only shows 3 ports on the Equalogic.

If it's true that within a stack an interruption (broken connector/port) will cause 30 to 60 seconds downtime i'm not to happy about it. We have 2 6224 switches which were advised by Dell. We have the stacking modules but we are not sure to use them.

Our setup is that an EqualLogic is connected to both switches (the issue as in question #6) and that both switches are connected to our blade-center and to main switch (HP 5406). This is to be able to route/connect to our other server-rooms. Most traffic is from the blade-center over the 6224's to the EqualLogic 6000.

If i would use a trunk ( Link Aggrigation) between the switches I will get a loop. But with RSTP I would expect downtime for maybe a couple hundred miliseconds. Which is far better than the switch stack I think. Could you tell what you opinion is on that, or maybe I'm overlooking something.

With the latest firmware there is no longer a 30-60 seconds downtime when the stack fails. That feature alone should be worth a Dell TechCenter blogpost with some video.

I agree, a quick test showed no downtime.

To answer my other question/ the question of dcu298 about how to connect the network cables. Just look at the large "quick start" paper that comes with the device. Controller A - 2 cables to switch A, 2 to switch B. With controller B the same.