FCoE not playing nice with our servers

[Pete89]

Hello,

About a year and a halp ago we had to make a desision about where we wanted our storage strategy to go. We had very little native FC so a move to something else would be pretty easy. Cisco got involved and pointed out that we would be in good position to take advantage of FCoE since our new data centers were full of NEXUS 7010s, 5548s, 2232s, and NetApp 3270s.

We're using Emulex OCE11102-FX cards, and HP ProLiant DL360 G7 servers. It seems Windows has a harder time. Windows sees SRB storage errors, and bouncing the ports in Emulex's software makes them go away. This can happen at any time and at times of the day when it can cause havoc.

On the LINUX systems we see:

rport-3:0-37: blocked FC remote port time out: removing rport

rport-5:0-43: blocked FC remote port time out: removing rport

Things have gotten so bad we are starting to look at jumping off the FCoE wagon despite having invested hundreds of thousands of dollars in the technology.

I guess the reason I started this thread was to see if anyone out there has seen this as well, and to know how well or not their FCoE implementation has gone.

Thanks for any advice or input,

P.

[vdsudame]

Hello

The problem you are reporting does not seem specific to FCoE as such but more to do with either Server or Storage problem. Do you have a TAC Case already opened with Cisco TAC and if so, please pass us the TAC Case number, or if not, can you open one ASAP so we can start investigating the nexus 5k switch logs? We might at some point also need to have a joint collaboration from HP/Netapp to get to the bottom of this issue.

Please let us know how we can assist.

Thanks, Vinayak

Technical Lead

Data Center Team

Cisco TAC.