You are not logged in.
I reinstalled Arch on my ThinkPad last month after some playing around with FreeBSD. It was running fine for a while, but around the time of the big glibc update it started giving me some error messages about eth0 at startup. I ignored it for a week or so because I normally connect over wireless, but now I've had some time to puzzle over it and I can't figure it out.
% dmesg | grep e1000e
[ 6.124136] e1000e: Intel(R) PRO/1000 Network Driver - 1.9.5-k
[ 6.124139] e1000e: Copyright(c) 1999 - 2012 Intel Corporation.
[ 6.124181] e1000e 0000:00:19.0: setting latency timer to 64
[ 6.124290] e1000e 0000:00:19.0: (unregistered net_device): Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode
[ 6.124339] e1000e 0000:00:19.0: irq 46 for MSI/MSI-X
[ 6.640922] e1000e 0000:00:19.0: (unregistered net_device): Hardware Error
[ 6.641176] e1000e 0000:00:19.0: eth0: (PCI Express:2.5GT/s:Width x1) 00:1a:6b:3a:2b:a9
[ 6.641180] e1000e 0000:00:19.0: eth0: Intel(R) PRO/1000 Network Connection
[ 6.641206] e1000e 0000:00:19.0: eth0: MAC: 6, PHY: 6, PBA No: FFFFFF-0FF
[ 43.330714] e1000e 0000:00:19.0: eth0: Hardware Error
[ 43.879375] e1000e 0000:00:19.0: irq 46 for MSI/MSI-X
[ 43.980255] e1000e 0000:00:19.0: irq 46 for MSI/MSI-X
[ 48.520186] e1000e 0000:00:19.0: eth0: Hardware Error
[ 48.596877] e1000e 0000:00:19.0: eth0: Hardware Error
[ 48.617178] e1000e 0000:00:19.0: irq 46 for MSI/MSI-X
[ 48.720110] e1000e 0000:00:19.0: irq 46 for MSI/MSI-XWhen I try to connect with it (using wicd), envision 24 more lines of alternating "irq 46 for MSI/MSI-X" and "eth0: Hardware Error". But I can connect, and I'm using it right now to post this as well as install the latest update. It doesn't seem to cause errors during normal usage.
I Googled a bit and searched the forums but didn't come up with anything that sounded like this issue. "Hardware Error" isn't a very descriptive message.
Edit - forgot to mention, both LEDs on the physical Ethernet port come on and stay on as long as I'm using wireless, but when I connect it to a cable they behave normally.
Last edited by Trent (2012-07-21 22:45:09)
Offline
I have a very similar problem on my Thinkpad T420. I installed Arch last weekend so I've always been post-glibc update.
I use wicd, and while wireless works great, ethernet doesn't. Even if I "manually" connect without wicd, it still doesn't work. I found a work-around, and that is to reload the module:
rmmod e1000e && modprobe e1000eAfter I reload, no hardware errors show up in dmesg and I can connect via ethernet both with and without wicd.
Offline
Mine's a T61.
Reloading the module doesn't change anything for me, that was one of the first things I tried. Maybe it's really a hardware issue. I'll have to boot a live system and see if it does the same thing.
Offline
Same problem with Acer Travelmate.
[ 10.023782] e1000e: Intel(R) PRO/1000 Network Driver - 1.9.5-k
[ 10.023785] e1000e: Copyright(c) 1999 - 2012 Intel Corporation.
[ 10.023847] e1000e 0000:00:19.0: setting latency timer to 64
[ 10.023984] e1000e 0000:00:19.0: (unregistered net_device): Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode
[ 10.024058] e1000e 0000:00:19.0: irq 47 for MSI/MSI-X
[ 10.352267] e1000e 0000:00:19.0: eth0: (PCI Express:2.5GT/s:Width x1) 00:a0:d1:a3:79:4f
[ 10.352271] e1000e 0000:00:19.0: eth0: Intel(R) PRO/1000 Network Connection
[ 10.352305] e1000e 0000:00:19.0: eth0: MAC: 6, PHY: 6, PBA No: FFFFFF-0FF
[ 25.416041] e1000e 0000:00:19.0: eth0: Hardware Error
[ 26.086491] e1000e 0000:00:19.0: eth0: Hardware Error
[ 26.709468] e1000e 0000:00:19.0: irq 47 for MSI/MSI-X
[ 26.810232] e1000e 0000:00:19.0: irq 47 for MSI/MSI-X
[ 152.210512] e1000e 0000:00:19.0: irq 47 for MSI/MSI-X
[ 152.976333] e1000e 0000:00:19.0: irq 47 for MSI/MSI-X
[ 153.077304] e1000e 0000:00:19.0: irq 47 for MSI/MSI-X
[ 396.970348] e1000e 0000:00:19.0: irq 47 for MSI/MSI-XOffline
I mentioned that I had to reload the e1000e module, but apparently I only needed to do that once. Every boot since then, networking works fine, but I still get these errors:
$ dmesg | grep e1000e
[ 28.292975] e1000e: Intel(R) PRO/1000 Network Driver - 1.9.5-k
[ 28.292984] e1000e: Copyright(c) 1999 - 2012 Intel Corporation.
[ 28.293095] e1000e 0000:00:19.0: setting latency timer to 64
[ 28.293386] e1000e 0000:00:19.0: (unregistered net_device): Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode
[ 28.293478] e1000e 0000:00:19.0: irq 43 for MSI/MSI-X
[ 28.539302] e1000e 0000:00:19.0: eth0: (PCI Express:2.5GT/s:Width x1) 00:21:cc:c8:04:b2
[ 28.539312] e1000e 0000:00:19.0: eth0: Intel(R) PRO/1000 Network Connection
[ 28.539367] e1000e 0000:00:19.0: eth0: MAC: 10, PHY: 11, PBA No: 1000FF-0FF
[ 39.223280] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 40.022622] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 40.822004] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 41.621410] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 42.420803] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 43.220185] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 44.022341] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 44.821769] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 45.621182] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 46.420986] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 47.220426] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 48.019865] e1000e 0000:00:19.0: eth0: Error reading PHY register
[ 48.460859] e1000e 0000:00:19.0: eth0: Hardware Error
[ 48.900153] e1000e 0000:00:19.0: eth0: Hardware Error
[ 49.045543] e1000e 0000:00:19.0: irq 43 for MSI/MSI-X
[ 49.151911] e1000e 0000:00:19.0: irq 43 for MSI/MSI-X
[ 52.291552] e1000e 0000:00:19.0: irq 43 for MSI/MSI-X
[ 55.205068] e1000e 0000:00:19.0: irq 43 for MSI/MSI-X
[ 55.307935] e1000e 0000:00:19.0: irq 43 for MSI/MSI-X
[ 58.144921] e1000e 0000:00:19.0: irq 43 for MSI/MSI-X
[ 60.512830] e1000e: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: None
[ 60.512843] e1000e 0000:00:19.0: eth0: 10/100 speed: disabling TSO(Edit: fixed bbcode tags)
Last edited by richli (2012-07-31 21:09:49)
Offline
Funny thing, mine stopped doing it today. The last adminy thing I did was install aisleriot and I doubt my Ethernet card was complaining about a lack of solitaire card games.
Weird, but I don't think I'll get any closer to an answer unless it starts happening again, so... oh well
Offline
I still get the irq stuff in my logs, but I've also noticed that it's really difficult to get an IP address from a DHCP server lately. I'm using wicd, by the way. Everything is fine with wireless, it's just connecting via ethernet is finicky.
It seems that after it puts the interface up, it launches dhcpcd but with the odd ethernet hardware it takes 20 or seconds before the ethernet link is established. Since dhcpd has a default timeout of 30 seconds, this eats up dhcpd's time and then it times out early.
Here's an example from /var/log/everything.log:
Aug 23 17:19:27 localhost dhcpcd[11183]: version 5.6.0 starting
Aug 23 17:19:27 localhost dhcpcd[11183]: eth0: waiting for carrier
Aug 23 17:19:27 localhost kernel: [13431.691112] e1000e 0000:00:19.0: irq 42 for MSI/MSI-X
Aug 23 17:19:48 localhost dhcpcd[11183]: eth0: carrier acquired
Aug 23 17:19:48 localhost kernel: [13452.620177] e1000e: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: Rx/Tx
Aug 23 17:19:48 localhost kernel: [13452.620190] e1000e 0000:00:19.0: eth0: 10/100 speed: disabling TSO
Aug 23 17:19:48 localhost kernel: [13452.620775] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
Aug 23 17:19:48 localhost dhcpcd[11183]: eth0: sending IPv6 Router Solicitation
Aug 23 17:19:48 localhost dhcpcd[11183]: eth0: sendmsg: Cannot assign requested address
Aug 23 17:19:48 localhost dhcpcd[11183]: eth0: rebinding lease of 10.8.116.12
Aug 23 17:19:52 localhost dhcpcd[11183]: eth0: acknowledged 10.8.116.12 from 10.8.125.2
Aug 23 17:19:52 localhost dhcpcd[11183]: eth0: checking for 10.8.116.12
Aug 23 17:19:52 localhost dhcpcd[11183]: eth0: sending IPv6 Router Solicitation
Aug 23 17:19:56 localhost dhcpcd[11183]: eth0: sending IPv6 Router Solicitation
Aug 23 17:19:57 localhost dhcpcd[11183]: timed outAs soon as the ethernet link is established ("carrier acquired"), it gets the IP address from the server. Then it checks if the address is in use on the network and appears to reach the 30-second timeout during that process.
I resolved the problem by editing /etc/dhcpcd.conf to increase the timeout by adding this line:
timeout 60And here is my log after increasing the timeout:
Aug 23 17:24:16 localhost dhcpcd[11787]: version 5.6.0 starting
Aug 23 17:24:16 localhost dhcpcd[11787]: eth0: waiting for carrier
Aug 23 17:24:17 localhost kernel: [13720.702559] e1000e 0000:00:19.0: irq 42 for MSI/MSI-X
Aug 23 17:24:38 localhost dhcpcd[11787]: eth0: carrier acquired
Aug 23 17:24:38 localhost kernel: [13741.744712] e1000e: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: Rx/Tx
Aug 23 17:24:38 localhost kernel: [13741.744725] e1000e 0000:00:19.0: eth0: 10/100 speed: disabling TSO
Aug 23 17:24:38 localhost kernel: [13741.746202] IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready
Aug 23 17:24:38 localhost dhcpcd[11787]: eth0: sending IPv6 Router Solicitation
Aug 23 17:24:38 localhost dhcpcd[11787]: eth0: sendmsg: Cannot assign requested address
Aug 23 17:24:38 localhost dhcpcd[11787]: eth0: rebinding lease of 10.8.116.12
Aug 23 17:24:41 localhost dhcpcd[11787]: eth0: acknowledged 10.8.116.12 from 10.8.125.2
Aug 23 17:24:41 localhost dhcpcd[11787]: eth0: checking for 10.8.116.12
Aug 23 17:24:42 localhost dhcpcd[11787]: eth0: sending IPv6 Router Solicitation
Aug 23 17:24:46 localhost dhcpcd[11787]: eth0: leased 10.8.116.12 for 3600 seconds
Aug 23 17:24:46 localhost dhcpcd[11787]: forked to background, child pid 11819It takes 22 seconds to acquire the ethernet carrier after dhcpcd starts up. Then it needs 8 more seconds before it finishes and forks to background. Having a 30-second timeout is too close, but the 60-second timeout works great.
Offline