[OmniOS-discuss] r151014 feedback

[Dan McDonald]

> On Apr 22, 2015, at 5:22 PM, Doug Hughes <doug at will.to> wrote:
> 
> The awesome: the nlockmgr bug appears to be fixed. We're not seeing that any more and nlockmgr is, so far, starting flawlessly.

Good.

> But, our brand new Intel 10G card is no longer recognized.. It started out ok, then a whole host of SFP+ not recognized over the course of about 1-2 minutes, and now the card will not attach the driver. We also had a whole lot of fmadm faulty at the same time. I acquitted them assuming they were spurious, but now dladm show-phys won't even see the ixgbe devices at all.
> 
> The summary says "
> 
> Fault class : fault.io.pciex.device-interr
> Affects     : dev:////pci@0,0/pci8086,340e@7/pci111d,806e@0/pci111d,806e@4/pci8086,7a11@0
>                   out of service, but associated components no longer faulty
> FRU         : "PCIe 1" (hc://:product-id=SUN-FIRE-X4275-SERVER:server-id=x4275-3-15-39:chassis-id=0931XFG005/motherboard=0/hostbridge=4/pciexrc=4/pciexbus=25/pciexdev=0/pciexfn=0/pciexbus=26/pciexdev=4/pciexfn=0/pciexbus=33/pciexdev=0)
>                   acquitted
> 
> 
> Is there a handy way to reset all of this and bring the device back into service and reset any notion of the faults?

Please look at your /var/adm/messages* files for ixgbe complaints.  There may be more useful data in there.  ixgbe generally is VERY VERY picky about which SFP+ one uses.

Dan