Brocade Fabric Assist Zones

Print Friendly, PDF & Email

Huh, what did they come up with now??? A new way of zoning? FCoE zones? Is this the opposite of target initiated zoning??

Well, no, actually nothing of such sort. Brocade abandoned FA and QL zoning over a decade ago but so very rarely I run into it. From a FOS configuration perspective these zone still operate but no management application is able to handle it any more.

The Brocade FA zones were mostly used in public-loop infrastructure where FL-ports serviced multiple public loops and in conjunction with QuickLoop zones. This way you were able to create special zones for loop enabled devices sitting behind these FL-ports and allowing private (ie direct attached hosts) to communicate with these loop devices. It also prevented certain loop maintenance tasks to have a negative impact on single exchange operations (i.e. for instance tape-backups). Brocade removed this function with FOS 4.1 from a CLI and Webtools perspective however FOS itself still adheres to the configuration. ever since when FOS 4.1 came out in around 2002/2003 the FCAL protocol has been on a decline as switches architectures were more and more affordable as well as required for higher performance and high-availability datacentres.

The FA zones had a special notation where an alias for an FA-zone was created in the form of H{20:00:00:00:c9:3f:e1:7a}.

If you have an environment (even up to FOS 7.3) and the zone-configuration has been propagated through a decade of hardware and software upgrades and you decide to install BNA you might run into a screen like this when you want to do a zoning configuration:

Selection_012(Yes, I ran into this in my own lab. :-))

If you’re then not proficient on the CLI and your webtools also plays up you might have a problem.

Anyway, the best way to circumvent this is to add the WWN as a separate entity to the alias and then remove the H{xx:xx:xx:xx:xx:xx:xx:xx} notation like this:

Sydney_ILAB_DCX-4S_LS128:FID128:admin> aliadd HSCTRN04,20:00:00:00:c9:23:6b:4e
Sydney_ILAB_DCX-4S_LS128:FID128:admin> aliremove HSCTRN04,H{20:00:00:00:c9:23:6b:4e}
Sydney_ILAB_DCX-4S_LS128:FID128:admin> aliadd train9500_e16c,20:00:00:00:c9:3f:e1:6c
Sydney_ILAB_DCX-4S_LS128:FID128:admin> aliremove train9500_e16c,H{20:00:00:00:c9:3f:e1:6c}
Sydney_ILAB_DCX-4S_LS128:FID128:admin> aliadd train9500_e17a,20:00:00:00:c9:3f:e1:7a
Sydney_ILAB_DCX-4S_LS128:FID128:admin> aliremove train9500_e17a,H{20:00:00:00:c9:3f:e1:7a}

After that do a cfgsave and cfgenable and wait till BNA has done a new inventory poll and you should be good to go.

Hope it helps in the very slight chance you run into this.




About Erwin van Londen

Master Technical Analyst at Hitachi Data Systems
Brocade , , , , , ,

Comments are closed.