Category Archives: hardware - Page 5

Troubleshooting connectivity issues on a Brocade SAN

Fog

I recently had the “pleasure” to figure out what was wrong with a Brocade based SAN environment. Servers were loosing connectivity on one of the HBAs, but all links were online and further investigation was necessary.

Going through all the error counters on each of the long wave SFPs finally revealed one of the SFPs’ health as marginal (hence it was still online, but very buggy indeed). The webtools GUI showed this particular SFP als orange instead of green. Disabling and re-enabling this SFP didn’t help and I decided to shut this SFP for good. And guess what: all my troubles went away. The trunk this SFP was in went back to a non-redundant, but healthy state and all servers got back to normal operations and got their redundant paths back.

So to summarize the story: look for marginal or even faulted SFPs when vague connectivity issues arise. If links are redundant, shutting the faulty one might help.

Adding or replacing a Cisco SAN switch in an IVR topology

Cisco MDS

If you have multiple datacenters or a multi tenant fibre channel environment and you’re using Cisco FC switches, it’s a best practice to use VSANs to separate the configurations of each location / tenant. To allow storage arrays and / or hosts in different VSANs to communicate with each other Inter VSAN Routing needs to be used.

If you need to have 2 EMC VNX storage arrays “talk” to each other for MirrorView for example over 2 or more datacenters (for data replication purposes that is) or hosts in one DC talk to storage in another DC, using transit VSANs (and therefore IVR) will keep your VSANs with equipment indoors and the slightly more vulnerable VSAN outdoors. If some farmer with his tractor rips your single mode fiber, only the outdoor VSAN will be fractured and the indoor VSANs remain unharmed. And of course communication between the remote sites is interrupted, but the indoor VSANs / fabrics remain unchanged.

Read more »

Cisco MDS NX-OS zoning on the CLI

Cisco MDS

The Cisco Fabric Manager or the newer DCNM can be great for creating a limited number of zones and aliases, but when the number of zones exceeds 10 or 20 or so, creating, cloning and editing these can be a pain in the B@TT.

FC-SWITCH-01# conf t
Enter configuration commands, one per line. End with CNTL/Z.

FC-SWITCH-01(config)#

Create aliases:

FC-SWITCH-01(config)# fcalias name VM101-HBA1 vsan 45
FC-SWITCH-01(config)# member pwwn 20:21:22:25:B6:00:00:01

Enhanced zone session has been created. Please ‘commit’ the changes when done.

FC-SWITCH-01(config-fcalias)#

FC-SWITCH-01(config-fcalias)# fcalias name VM101-HBA1 vsan 45
FC-SWITCH-01(config-fcalias)# member pwwn 20:21:22:25:B6:00:00:02
FC-SWITCH-01(config-fcalias)# fcalias name VM102-HBA1 vsan 45
FC-SWITCH-01(config-fcalias)# member pwwn 20:21:22:25:B6:00:00:03

FC-SWITCH-01(config-fcalias)# device-alias commit

Create two new zones:

FC-SWITCH-01(config-zone)# zone name VM101-HBA1_VNX-5600-08-SPA6 vsan 45
FC-SWITCH-01(config-zone)# member fcalias VNX5600-08-SPA6
FC-SWITCH-01(config-zone)# member fcalias VM101-HBA1
FC-SWITCH-01(config-zone)# zone name VM102-HBA1_VNX-5600-08-SPB7 vsan 45
FC-SWITCH-01(config-zone)# member fcalias VNX5600-08-SPB7
FC-SWITCH-01(config-zone)# member fcalias VM102-HBA1
FC-SWITCH-01(config-zone)# zone commit vsan 45

Commit operation initiated. Check zone status

Now add the two new zones to a zoneset:

FC-SWITCH-01(config)# zoneset name ZS_VSAN170 vsan 45
FC-SWITCH-01(config-zoneset)# member VM101-HBA1_VNX-5600-08-SPA6
FC-SWITCH-01(config-zoneset)# member VM102-HBA1_VNX-5600-08-SPB7
FC-SWITCH-01(config-zoneset)# zone commit vsan 45

FC-SWITCH-01(config)#

Maximum distances using fiber cable types OM1/OM2/OM3/OM4/OS1

What are the maximum distances using different fiber grades?

 

fiber cable specs graph

The graphs show the direction where OM-specs are going compared to distances.

Read more »

Cisco MDS: Fabric is already locked

Cisco MDS

When you encounter a fabric lock, because you accidentally left the GUI or CLI without committing the changes, you can try the following to clear the lock and retry to apply your changes:

  1. run ‘show cfs lock’ to see who lock`s the fabric
  2. run ‘clear device-alias session’ to clear the lock when you were doing zoning activities

Instead of the “clear device-alias session” in line 2, another common possibility to get the lock cleared is:

  • ‘clear ivr session’ (when you were in the middle of IVR activities)

Other locks can occur, but the device-alias and ivr are probably the most common. At least the ones that I encountered so far.

Sometimes a lock error pops up when opening a VSAN in DCNM, saying that the fabric is locked from switch ABC. In that case, log on to switch ABC and type (if nobody is actually zoning at that moment):

  • clear zone lock vsan 123

This will clear the lock from the entire fabric only if issued on initiating switch