Basic NDMP connectivity question

aux0 should automatically login, i've never had to force anything on the datamover side. Safe to assume that port on the switch is not "shut" ?

yes, attaching a FC tape drive through a FC switch to the AUX0 port is the most common config

There aren't many config options there - see the NDMP manual

I think you need to run a server_devconfig to discover the tape drives

What is your switch vendor and model?  As dynamox says, be sure the port is not shutdown.  You shouldn't have to set anything on the switch port.  Can you see anything in the switch logs, like a portfail or negotiation error?  Twice now, I've used a bad patch cable on the AUX0 and AUX1 ports, then wondered why I had no link.  Otherwise, you can use the server_devconfig commands to try and scan the SCSI bus for devices.

server_devconfig server_# -probe -scsi -nondisks -all should scan all available buses (including the AUX0 and AUX1 ports) without having to reboot the datamovers.  You should see some kind of actvitity on the switch port, perhaps even some errors, to help debug any switch errors.

Let us know how it goes!

Karl

I agree with with Karl - for this kind of problem it's rarely the Celerra - most likely cabling or switch (or plugging the cable into the wrong or standby data mover )

Rainer

Guys,

Thankyou all for you comments.

It would appear after all that I have a duff AUX0 port on the primary data mover. I get an instant link up and fabric login if I connect either AUX1 on the primary, or AUX0 on the standby.

Regards

Andrew.

try swapping the SFP's - most likely it's just the SFP being dead or not fully plugged in

they are customer replaceable online

Yeah – I did try that first.

In hindsight I should have tried the other ports before posting the question, but you don’t expect a new onboard port to be dead. I assumed I must have been missing something.

Good morning

You can try running a .server_config command against the port.  Below are the ones that may help get the port working again:

fcp portreset=n ........ Reset fibre port n
fcp rediscover=n ....... Force fabric discovery process on port n Bounces the link, but does not reset the port
fcp rescan ............. Force a rescan of all LUNS
fcp reset .............. Reset all fibre ports

The command would be similar to the following:
.server_config server_2 -v "fcp portreset=2” [forcing HBA2 to log back into fabric]

Also, please note that support may request a datamover reboot before attempting to have the current one replaced.  If you decide to that, the the above are not required.

Let me know if this helped.

Sebby Robles
eServices Support
EMC Celerra Support