05-07-2010 03:43 AM
I'll outline the problem we're having..
We have two 200E's linked together via on a 4G port, they're right next to each other there isnt any long distance stuff involved. There's no errors zoning is setup correctly and both switches have a full fabric license.
We run VMware on all hosts attached the switches.
The situation is that traffic between the two switches is slow.
For example On switch 1 we have a Server that has its Hard-drive stored on the SAN attached to Switch number 2. On Startup this server takes around 5 minutes to startup plus copy a 500MB folder its estimates that it would take 20mins to copy onto its harddrive.
If I move the Server to Switch2 the startup then only takes 30-40 seconds and the same folder copy takes under a minute.
There's no errors as far as I can see on the port.
The two switches are on different OS's but not that different any help would be appreciated.
Switch 1: OS 6.2.1
Switch 2: OS 6.2.2b
05-10-2010 02:24 AM
1. Are there any error on ESX when the server boots up via switch 1. If there are SCSI reservation issues, then you should see some related errors.
2. Are all other servers on switch 1 with storage on switch 2 having the same problem.
3. Provide port number of ISL's on both switches and attach the o/p of fabriclog -s by pasting it in a txt file. (for easier interpretation).
05-14-2010 02:40 AM
Check if there is any errors in your switch error log (errshow, salhealth, etc).
Check if your long distance mode is configured correctly to L0.
Set your ISL ports to a 4Gb by hand, no autodetect. Check that your FC cable in the ISL is in good condition.
Also check if your port is okay, and no errors or performance problems at the switch level. (portstatsshow, fabricwatch, perfmonitorshow, etc).
If that's okay then I would better point the finger at the Server patches, upgrade it, check your multipathing software, etc.
There are many things to check. If you solve the issue please write the answer back in this thread so we can also learn from it! =)