2x Stacked PowerConnect 6224 with fibre redundancy to 3rd 6224

Networking

Networking
Learn how to modernize & transform the network with the latest solutions from Dell Data Center, Campus and Wireless networking solutions.

2x Stacked PowerConnect 6224 with fibre redundancy to 3rd 6224

This question is answered

Hi,

 

I was wondering if someone could offer me some advice on the best way to setup 2x stacked PowerConnect 6224s with fibre redundancy to a 3rd PC6224?

 

Basically we have an HQ site with two stacked switches to provide redundancy to our servers/SANs. We also have a DR site with a single PC6224 with which we wish two attach two 10Gb fibre links to for path redundancy to a the passive node of a server cluster and replicated SAN.

 

How would it be best to configure the fibre links on the switches? Would I need to set up link aggregation on PC1/x23 & PC2/x24 to go to PC3/x23 & x24. Or would I be best to just let spanning tree sort it out?

 

Thanks for any help,

 

James.

Verified Answer
  • I guess I'm a little confused.  Sorry if I'm overstating the obvious.

    When 2 or more switches are stacked they act as if they were one single switch with more ports physical ports available.  So there is no real set up between the stacked switches (other than the stack set up).  

    As far as connecting to the third 6224 non stacked it is best practice to set up a LAG.

    Get Support on Twitter @DellCaresPro

    Download the Dell Quick Resource Locator app today to access PowerEdge support content on your mobile device! (iOS, Android, Windows)

All Replies
  • Link Aggregation will increase the link speed beyond the limits of any one single cable or port, and increase the redundancy for higher availability.

    The benefits of Link Aggregation include:

    • Increased Availability — If a link within a LAG fails or is replaced, the traffic is not disrupted and communication is maintained (even though the available capacity is reduced).

    • Load Sharing — Traffic is distributed across multiple links, minimizing the probability that a single link be overwhelmed.

    • Use of Existing Hardware — Firmware replaces the need to upgrade the hardware to higher bandwidth capacity.

    channel-group

    Use the channel-group command in Interface Configuration mode to

    configure a port-to-port channel. To remove the channel-group configuration

    from the interface, use the no form of this command.

    Syntax

    channel-group port-channel-number mode {on|auto}

    no channel-group

    • port-channel-number — Number of a valid port-channel for the current port to join.

    • on — Forces the port to join a channel without LACP.

    • auto — Forces the port to join a channel with LACP.

    LACP

    LACP provides a standardized means for dynamically exchanging information between two switches in order to configure and maintain LAGs automatically. LACP can automatically detect the presence of other aggregation-capable network devices in the system. It enables the user to determine which links in a system can be aggregated. For each aggregatable link, the switches exchange LACP frames for synchronizing the databases of the connected LACP enabled ports. In order to group the ports to a Link Aggregation group, the LACP analyzes the database periodically to detect LACP enabled ports.

    Hope this helps,

    Get Support on Twitter @DellCaresPro

    Download the Dell Quick Resource Locator app today to access PowerEdge support content on your mobile device! (iOS, Android, Windows)

  • Hi,

    I have already seen this in the 6200 series config guide but I'm still a little unsure as it doesn't mention if this is possible across a stack? I presume it is just looking for clarification really.

    Also, it's probably worth mentioning that there are only 10x 1Gb connections in the DR switch so I don't think that we'll actually need the increased bandwidth that LAG provides (unless we expand our server farm but that prob won't be happening in the near future).

    Thanks for the advice,

    James.

  • I guess I'm a little confused.  Sorry if I'm overstating the obvious.

    When 2 or more switches are stacked they act as if they were one single switch with more ports physical ports available.  So there is no real set up between the stacked switches (other than the stack set up).  

    As far as connecting to the third 6224 non stacked it is best practice to set up a LAG.

    Get Support on Twitter @DellCaresPro

    Download the Dell Quick Resource Locator app today to access PowerEdge support content on your mobile device! (iOS, Android, Windows)

  • I did read that they acted as a single switch but as I am setting this up for the primary reason of redundancy/HA and the extra bandwidth/load balancing provided by the LAG is not required, the reasons for setting up a LAG would be pretty much redundant.

    But if it's best practice then that's the way I'll go.

    Thanks again.