Today I added two edge (9216A) switches to my core (9509) switches. Each edge and a core had one ISL in VSAN0001 created so the edge could become part of the single fabric. The core switches have two ISLs between them for management.
I changed the core to edge ISLs (2 off)into a Port Channel and created a VSAN across each core and edge switch using the Port Channel. There are no active interfaces in VSAN0001 in the edge switches. All works very well in the non VSAN0001 vsans..
During this change the principal switch decided to move to the other core switch (because I never set a priority on one to start with). It seems that when I introduced the two edge switches, something obviously made the principal change. I don't think this is a show stopping issue but I think I need to set the priority lower on one of the core switches to stop it changing in the future. I changed the priority of the edge switches after I noticed the principal change. I am assuming that setting the priority is dynamic and wont affect any of the production VSANs.
The problem is now that one 9509 has a VSAN0001 in segmented mode and both the 9216A switches have VSAN0001 in a down segmented mode.
My Cisco training discussed this but as per usual it was something that we intentionally did as an exercise and there was an easy way to fix it. I think the config and troubleshooting training needs to be extended to two weeks.
Could someone please advise on how to fix this issue. As it is in a non important VSAN, I am not really concerned right now.
I have a bunch of other switches to add to the fabric and don't really want to go through this again.
The preference for deciding principal is 1.) number of domains already known. If a switch already has a view of a SAN fabric, then it is best to have it win PS. 2.) priority 3.) SWWN
So, if you look in the log, it will tell you why it segmented as well as in the show interface. Domains ID's will cause segmentation only in situations where a particular DID is required such as interop or if the DID's are overlapping and statically configured. Other than in a scenerio where specific DID's are required or overlappng, then it will be a zoneset mismatch or beyond the domain assignment.
So, find out what it is exactly what caused your segmentation. You can post it here from the log or the show interface if you like and we can try to work thru it.
I had a very good look at the logs and could not find anything of value. Thats a bit odd to say the least.
Each Domain ID is unique as I have statically assigned them.
Considering I don't have any zonesets/zones in VSAN0001, this whole issue is mystifying to me. Normally zoning issues are the cause of segmentation or so I believe.
The only thing that I can think of is that I created one ISL between each of the core and edge switches and it used VSAN0001. When I added the second ISL and then made both into a truncking port channel where only one VSAN was allowed (eg, 500 on one core/edge and 501 on the other core/edge), this might have broke the management connection between the switches and thus segmented the VSAN0001.
I think that happened the first time I added two switches together and was promptly informed to reverse that (and then I put it back after I convinced our projects people I knew what I was doing). When I turned off the ISL, I got a segmented VSAN in 0001. When I put them back together, it fixed itself.
Actually my theory sounds good to me and I thought I might be able to fix it by allowing VSAN0001 to use the port channel. Problem is I could not work out how to do it. Fabric Manager would not allow me (perhaps through my own stupidity) to add VSAN0001 to the allowed VSAN's the port channel could use.
When I get to work on Monday, we will have more fibre cabling and I am interested in seeing if I add another ISL between the switches and let it use VSAN001, this should sort itself out naturally.
I know that you can have many VSANs through a port channel so I wonder how to make the change.
I would prefer not to have to use an ISL between the switches if I can use a Port Channel as well.
I finally figured out how to allow more VSAN's through the port channel. It can be done using the Fabric Manager.
So once I let VSAN0001 through the port channel, the two segmented VSAN0001's on the edge switched disappeared. But I still had a segmented VSAN0001 on the principal switch. I finally figured out VSAN0001 on an edge switch had a preferred Domain ID that was identical to one of the static Domain ID's on another VSAN on the other core switch. I changed the Domain ID on the edge switch and everything is back to normal.
I never thought about assigning static Domain ID's to VSAN0001 as I would never create a zoneset/zone in them.
The unmanaged mode is also known as Network only switching, which is introduced in Brazos release. It adds the flexibility for customer to use only network automation for service appliance.
If a device is configured a...
Usually, we can access ESXi Shell by pressing Alt+F1 from ESXi DCUI (Direct Console User Interface).
But on HyperFlex system, it just shows black window.
This is expected behavior because HyperFlex redirects ESXi Shell output to SoL...
Configuring an Export Policy Using the GUI
This procedure explains how to configure an Export policy using the APIC GUI. Follow these steps to trigger a backup of your data:
On the menu bar, choose Admi...