Bug 32662

Summary:	r8169: eth0: link is not ready
Product:	Drivers	Reporter:	pjwncmfmzxg
Component:	Network	Assignee:	Francois Romieu (romieu)
Status:	RESOLVED OBSOLETE
Severity:	normal	CC:	alan, nipsy, romieu
Priority:	P1
Hardware:	All
OS:	Linux
Kernel Version:	3.0	Subsystem:
Regression:	No	Bisected commit-id:
Attachments:	uname and dmesg output

Description pjwncmfmzxg 2011-04-04 23:28:28 UTC

Greetings,

with my Lenovo laptop and:
Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast Ethernet controller (rev 02)
with the r8169 module I always get:
"
r8169: eth0: link down
ADDRCONF(NETDEV_UP): eth0: link is not ready
"
when I load the module and after that any action with eth0 won't work.

I've tried reloading the module or disabling auto negotiation a la:
http://adam.rosi-kessel.org/weblog/2008/06/21/a-much-simpler-fix-for-the-r8169-link-down-problem and in various other bug reports, but nothing does it.

Any help on fixing this problem is greatly appreciated.

Comment 1 Francois Romieu 2011-08-04 09:10:34 UTC

Do you still notice the same behaviour with a recent post 3.0 kernel w/o ACPI
disabled ?

Is so a complete dmesg would be welcome.

-- 
Ueimor

Comment 2 Mark Nipper 2011-08-25 03:29:17 UTC

I'm adding an attachment of this happening with a 3.0.0 kernel (Debian testing/unstable kernel).

Comment 3 Mark Nipper 2011-08-25 03:29:49 UTC

Created attachment 70132 [details]
uname and dmesg output

Comment 4 Francois Romieu 2011-08-25 07:19:11 UTC

(In reply to comment #3)

It's a different 816x chipset. The symptoms look close.

The PME# weirdness does not seem to appear at module insertion time but later.

Does it appear randomly while using the computer (or while going to lunch)
or does it vaguely correlate with any kind of event ?

Thanks.

-- 
Ueimor

Comment 5 Mark Nipper 2011-08-26 02:20:21 UTC

Nothing specific.  This is on my Linux file server at home and it usually just sits around doing a lot of nothing unless we're watching something from it.  I have some background processes running which should be keeping the NIC active pretty much all the time assuming that it has anything to do with inactivity.  But it's also possible that we could have experienced some downtime via our ISP which caused the NIC to be less active than normal if it does in fact have anything to do with activity.

Let me know if you want me to do add in any debugging code or anything else which might be helpful.

Comment 6 Alan 2015-02-19 16:22:03 UTC

This bug relates to a very old kernel. Closing as obsolete.