[Bug 229392] Re: (net 8139too) system with two of these ethernet cards, under any reasonable load, generates NETDEV WATCHDOG: eth0: transmit timed out
Ixhausen
harneit at physik.fu-berlin.de
Mon Aug 24 00:30:20 UTC 2009
Aaaah! I heff ze internet now!
The (trivial!) solution to the single-ethernet-adapter branch of the bug
above was actually in
http://ubuntuforums.org/showthread.php?t=538448
Briefly, my RealTek adapter was just put to sleep by Windows on
shutdown, and kubuntu could not re-awaken it (even in kernel
2.6.31-999).
I simply followed the advice given at the link above and told WinXP to
leave the RealTek on (by starting the device manager, locating the
RealTek adapter there, dbl-clicking it, and applying the following
setting in the advanced tab: Wake-On-Lan After Shutdown = Enabled).
This behavior may be more general than I thought before. Lots of new
*ubuntu users complain about very bad network support, which is of
course a show-stopper for the transitionally-minded.
Maybe the Wake-On-Lan feature is now ubiquitous in the ethernet adapter
world?
I guess that many newbies actually start off with a dual-boot
installation, or just try the liveCD RIGHT AFTER WINDOWS SHUTDOWN (i.e.
possibly with disabled network devices). The work-around is REALLY easy.
But as an ubuntu/kernel/network bug, almost impossible to trace.
Maybe this hint (enable Wake-On-Lan if you experience network problems and if you're a double-booter, or try the liveCD from a windows system), could be included in the "known issues" section of the distro???
Whom do I tell?
What to do with this bug now? At least my comments and the one by IKT
are probably misfiled, at least I imagine that the sporadic malfunction
of a double ethernet setup is due to other problems...
--
(net 8139too) system with two of these ethernet cards, under any reasonable load, generates NETDEV WATCHDOG: eth0: transmit timed out
https://bugs.launchpad.net/bugs/229392
You received this bug notification because you are a member of Kernel
Bugs, which is subscribed to Linux.
More information about the kernel-bugs
mailing list