> 
>>> If these are the problem, try using just one (igb0) for starters and 
>>> "ifconfig igbX down" for 1, 2, and 3.
> 
> That is exactly what i did.. for each test brought the others down 
> Just to add, when doing a  SH ARP on my cisco's, don't see the macs 
> popping up, which is even more worring , no arps on a switch usually 
> is very bad news

And you've eliminated the switch as a problem?  I ask because of this recent 
push (don't let the commit date fool you, this went in in November):

Yes... using the e1000g's and the same switch/etherchannel works flawless, 
also, using one of my vmware machines, and testing the same etherchannel works 
perfect.
Beside this is a cisco c3750 e, running 12.2 firmware, which is about as proven 
as the way to rome :-)

commit b607c8a3bdb27d4fde6e3fc4bb6617a1d91bdca7
Author: Keith M Wesolowski <[email protected]>
Date:   Tue Aug 5 22:08:11 2014 +0000

    5337 igb/ixgbe mishandle raw packets if cable problem
    Reviewed by: Robert Mustacchi <[email protected]>
    Reviewed by: Dan McDonald <[email protected]>
    Reviewed by: Jason King <[email protected]>
    Reviewed by: Garrett D'Amore <[email protected]>
    Approved by: Richard Lowe <[email protected]>

 usr/src/uts/common/io/igb/igb_tx.c     | 10 +++++++---
 usr/src/uts/common/io/ixgbe/ixgbe_tx.c |  6 ++++--
 2 files changed, 11 insertions(+), 5 deletions(-)

This is the ONLY change since the February of 2014 support for I354 that's gone 
into the Intel GigE NIC code.

>>> Also, I need to ask --> are either of your e1000g0's shared IPMI/host 
>>> links?  If so, disable the sharing feature in the BIOS.  We don't cope with 
>>> shared-with-IPMI NICs.
> 
> They are not... besides.. the e1000G's ( and their aggr) are working fully as 
> expected.
> To make it more interesting.. this was the same setup as I had on r11 
> , and then it did seem to work 100%

The only difference between 011 & 012 and 013 is the above commit.  You 
*could*, if you're feeling adventurous, swap out the "igb" driver in 013 with 
one from 012 or 011.  And if you unplumb all of the igbs (both in IPv4 and 
IPv6), you MAY be able to do it without rebooting.

As i can't use the machine for it's primairy function ( ISCSI target utilizing 
those i350's), rebooting it's any issue.
Just not sure as HOW to change that driver, so would be great if you'd have 
some pointers

Do you know which PCI ID your igb (it's an I350, right?) board is?  "prtconf -d 
| grep -i network" should be a quick way to show me that.

pci8086,1 (pciex8086,1521) [Intel Corporation I350 Gigabit Network Connection], 
instance #0
            pci8086,1 (pciex8086,1521) [Intel Corporation I350 Gigabit Network 
Connection], instance #1
            pci8086,1 (pciex8086,1521) [Intel Corporation I350 Gigabit Network 
Connection], instance #2
            pci8086,1 (pciex8086,1521) [Intel Corporation I350 Gigabit Network 
Connection], instance #3


floris
...:: House of Ancients ::...
American Staffordshire Terriers

+31-628-161-350
+31-614-198-389
Het Perk 48
4903 RB
Oosterhout
Netherlands
www.houseofancients.nl
_______________________________________________
OmniOS-discuss mailing list
[email protected]
http://lists.omniti.com/mailman/listinfo/omnios-discuss
  • [OmniOS-discuss] LACP... Floris van Essen ..:: House of Ancients Amstafs ::..
    • Re: [OmniOS-disc... Floris van Essen ..:: House of Ancients Amstafs ::..
      • Re: [OmniOS-... Dan McDonald
        • Re: [Omn... Floris van Essen ..:: House of Ancients Amstafs ::..
          • Re: ... Dan McDonald
            • ... Floris van Essen ..:: House of Ancients Amstafs ::..
              • ... Dan McDonald
                • ... Floris van Essen ..:: House of Ancients Amstafs ::..
                • ... Dan McDonald
                • ... Floris van Essen ..:: House of Ancients Amstafs ::..
                • ... Floris van Essen ..:: House of Ancients Amstafs ::..
                • ... Eric Sproul
                • ... Dan McDonald
                • ... Floris van Essen ..:: House of Ancients Amstafs ::..
                • ... Dan McDonald
                • ... Floris van Essen ..:: House of Ancients Amstafs ::..
                • ... Dan McDonald
                • ... Floris van Essen ..:: House of Ancients Amstafs ::..

Reply via email to