Extend Existing Fabric

[Anonymous]

I did not set the domain ID, an onsite engineer performed the configuration. The fabric.ops shows different IDs.

FCSW10:admin> config | grep fabric.domain
fabric.domain:10
FCSW10:admin>

=========================

FCSW12:FID128:admin> configshow | grep fabric.domain

fabric.domain:2
FCSW12:FID128:admin>

[Anonymous]

I'm not familiar with trunkdebug command so I'll have to do a little research. Is there anything I need to watch out for?

[Anonymous]

Please check the output on both switches of: defzone --show

Send also the error message for your ISL segmentation: errshow -r

Firmware version on both switches: firmwareshow

Post the output here for us to check.

[Anonymous]

Here is the requested output. I must apologize again, I was working on the assumption that the engineer upgraded the new switches to FOS 6.4.2 to match our existing switches.

FCSW10:admin>
FCSW10:admin> defzone --show
Default Zone Access Mode
committed - No Access
transaction - No Transaction

FCSW10:admin> errshow -r
Fabric OS: v6.4.2

2011/11/14-23:38:39, , 5657, FID 128, WARNING, FCSW10, port 45, domain IDs overlap.
2011/11/14-23:29:16, , 5656, FID 128, WARNING, FCSW10, port 45, domain IDs overlap.

FCSW10:admin> firmwareshow
Appl     Primary/Secondary Versions
------------------------------------------
FOS      v6.4.2
         v6.4.2
FCSW10:admin>

===================================================================

FCSW12:FID128:admin> defzone --show
Default Zone Access Mode
committed - No Access
transaction - No Transaction

FCSW12:FID128:admin> errshow -r
Fabric OS: v6.4.1a

2011/11/14-23:38:39, , 207, FID 128, WARNING, FCSW12, port 45, domain IDs overlap.
2011/11/14-23:29:16, , 206, FID 128, WARNING, FCSW12, port 45, domain IDs overlap.

FCSW12:FID128:admin> firmwareshow
Appl     Primary/Secondary Versions
------------------------------------------
FOS      v6.4.1a
         v6.4.1a

[Anonymous]

You have domian ID overlap.

You have in both fabrics the same Domain ID configured.

Do you have any FCR connected to or both switches which try to merge? you have to check which Domain IDs are configured.

Regards,

Andreas

[Anonymous]

Ok, I can see a few things. First of all, are you using virtual fabrics in any of your switches? Try this: fosexec --fid all -c "fabricshow"

In FOS <= 6.4 if you have policy=noaccess and no configuration enabled, this may cause problems when merging with a fabric with config enabled and policy=noaccess. Therefore you have to set policy = allaccess on the new switch before merging.

Do the following on FCSW12:

defzone --allaccess

cfgsave

Try merging once again.

If you still get the same message do the following on FCSW12:

configdefault

reboot

Try to merge once again.

If you run into errors once again then post the output of this command:

     fabstatsshow

[Anonymous]

Here's a complete picture of my environment.

I do have a couple of Brocade 7500s. It appears two of the new switches at Site B were configured with the same domain IDs as the two new switches at Site A. Obviously this is not an ideal configuration. However, the FCR at Site B is not physically connected to the new switches, so I don't understand how that conflict would be identified. Also, why would Fabric A at Site A not also have the same problem?

Regardless of the answers to those questions, it looks like my best course of action is to go ahead and upgrade the FOS on FCSW10 and FCSW12 to 6.4.2 as well as change the domain IDs. Would you agree?

Site A

Fabric A - FCSW09 and FCSW11

============================

FCSW09:admin> configshow | grep fabric.domain
fabric.domain:105

--------------------------------------------------------

FCSW11:FID128:admin> configshow | grep fabric.domain
fabric.domain:1

Fabric B - FCSW10 and FCSW12 < ----- ISL not working "(segmented)(domain overlap)"

============================

FCSW10:admin> configshow | grep fabric.domain
fabric.domain:10

--------------------------------------------------------

FCSW12:FID128:admin> configshow | grep fabric.domain
fabric.domain:2

FC7500x01 - FCR

============================

FC7500x01:admin> configshow | grep fabric.domain
fabric.domain:11

Site B

Fabric A - FCSW13 and FCSW15

============================

FCSW13:FID128:admin> configshow | grep fabric.domain
fabric.domain:1

--------------------------------------------------------

FCSW15:FID128:admin> configshow | grep fabric.domain
fabric.domain:3

Fabric B - FCSW14 and FCSW16

============================

FCSW14:FID128:admin> configshow | grep fabric.domain
fabric.domain:2

--------------------------------------------------------

FCSW16:FID128:admin> configshow | grep fabric.domain
fabric.domain:4

FC7500x02 - FCR

============================

FC7500x02:admin> configshow | grep fabric.domain
fabric.domain:12

[Anonymous]

Can you disable the switch FCSW12?

If so change the domain ID to another value like 99 and reenable the ISL and check wich message will be logged in the errdump.

You write this: Fabric B - FCSW10 and FCSW12 < ----- ISL not working "(segmented)(domain overlap)"

For my understanding does this mean you have the same Domain ID in you fabric. Keep inmind that you may have some XD  and FD ID in your fabric when router are in place.

Regards,
Andreas