[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: e1000e driver bug makes some net-installations fail.



On Mon, Sep 27, 2010 at 18:29:36 -0300, Otavio Salvador wrote:

> Hello Jon,
> 
> On Mon, Sep 27, 2010 at 12:42 PM, Jon Ander Ortiz <jonbaine@gmail.com> wrote:
> ...
> > This bug is explaned here, basically the ethttool interface of the e1000e
> > driver, returns "2" instead of "1" when the
> > driver is asked about the link and the interface has link.
> ...
> > The fix, can be done using a newer version of the installer, or patching the
> > installer kernel with the fix, or providing
> > a patched version of the netfg to the remote systems.  This is a problem of
> > a buggy driver, not of the d-i, but i think
> > that it's interesting to make it public in this list, and more since many
> > cards are being migrated from e1000 to e1000e driver
> >  (e.g. 82573L) and with a debian installer update the net installation
> > fails.
> 
> Please report a bug against linux-2.6 source package and detail this
> stuff there so people can work for including this fix in next stable
> update.
> 
FWIW:
18:01 < jcristau> <[🔎] AANLkTin1p9fK6Ofxz3U_JO6rbRwiBpLFnEy2gA_KT0R1@mail.gmail.com> sounds like it might be worth fixing in stable
18:05 < bwh^> Should be fixed in d-i; noone promises that ETHTOOL_GLINK yields 
              0 or 1
18:06 < bwh^> Simple test for the ethtool API: does ethtool uses that feature? 
              if not, don't assume it works.

Otavio, it seems this ought to be fixed on the d-i side?

Cheers,
Julien

Attachment: signature.asc
Description: Digital signature


Reply to: