On 16.06.2011 18:06, Ronciak, John wrote:
> We are talking about the igb driver and not e1000, correct?  Just
> want to make sure we are talking about the correct driver.  BTW, our
> very latest drivers can be found at http:/e1000.sf.net which has all
> of our latest drivers including the igb driver.  You might want to
> try our latest version.

Yes, it's the igb driver, sorry if I made any confusion here.


>> I signaled the issue before, and it was suggested I may need to
>> update network card's firmware:
>>
>> http://comments.gmane.org/gmane.linux.network.general/14930
> Who told you to update the firmware?  The link doesn't show anything
> like that.  We don't hand out updates to firmware.  It's possible
> that it can be updated via a system BIOS upgrade.  Is this system
> running the very latest system BIOS from HP?

I've applied all BIOS and other firmware updates, but it didn't change 
this behaviour.

FYI, although not really related to this list, HP ProLiant DL180 G6 has 
a bug in its "management firmware" (iLO), which causes very strange 
network behaviour after the server is online for some time:

- almost exactly every hour, packet delays ranging 10-20 seconds for 
about a minute (ICMP packets sent from a machine connected to the same 
switch).

I've observed this behaviour on ~8 such servers (at least, there was 
such a bug with hardware shipped 1.5 yrs ago); firmware update from HP 
fixes it.


>> http://virtall.com/files/temp/IMG00066-20110420-0956.jpg
>> http://virtall.com/files/temp/IMG00067-20110420-0957.jpg
>> http://virtall.com/files/temp/2011-06-15_16-19-51_252.jpg
> The all just show the stack, nothing about the driver.  What makes
> you think that it's the driver?

Perhaps the comment from the gmane thread (and, that the box is pushing 
a lot of traffic):

        Are you running the latest version of the NIC driver? It's more
        likely that the problem is caused by the driver then by the
        kernel.

        Although the Intel drivers are not known to be flaky, your load
        is pretty high (if it's on constantly on this level) and could
        reveal a rare race.


But you're right, there is nothing related to the driver in the output.

I'll try to get a serial console to see if I can capture some more data.


-- 
Tomasz Chmielewski
http://wpkg.org

------------------------------------------------------------------------------
EditLive Enterprise is the world's most technically advanced content
authoring tool. Experience the power of Track Changes, Inline Image
Editing and ensure content is compliant with Accessibility Checking.
http://p.sf.net/sfu/ephox-dev2dev
_______________________________________________
E1000-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel® Ethernet, visit 
http://communities.intel.com/community/wired

Reply via email to