There are a couple of different things you can look at. To see if things look 
good at the qlc driver layer you can enable extended logging in the 
/kernel/drv/qlc.conf and reboot and duplicate.

Then I'm not sure if these next two things were backported to update 2 of S10.

run MDB:

# mdb -k

gather the fp trace. if you get an error then these next two dcmds were not 
backported...
> ::fptrace

gather the fcptrace
> ::fcptrace

And then quit mdb:
> $q

So interpret those I suggest reading the "getting to know your san stack" 
presentation that can be found by googling it. You could also post those logs 
and the /var/adm/messages after enabling the qlc.conf extended logging.

You could also use the scsi.d dtrace script to trace all the SCSI I/O run 
through MPXIO to see what's going on. http://blogs.sun.com/chrisg/page/scsi.d

Are there any other messages in the /var/adm/messages surrounding the mpxio 
message?

What type of storage is this? Is it possible there is an internal maintenance 
routine within the storage that is occurring at this time? Snapshots perhaps?
 
 
This message posted from opensolaris.org
_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss

Reply via email to