{Solved] was Myth Network woes...Re: Local source Intel gigE NIC? / RTL8169 woes... was Myth / network question

[Bruce Labitt]

Fixed my network drop out issue on myth.  Issue was RTL8169 NIC dropping 
connection.
Left no meaningful trace in logs besides link down/up.  Network would 
restart but drop again in 10min - 3 hours.  Certainly not reliable for 
recording.

Replaced NIC with vanilla Intel NIC 1000Mbit.  No more problems.

Have reason to believe that it is directly related to the RTL8169 since 
I tried two different 8169 based NICs with same result.

YMMV

Have a good holiday everyone!

-Bruce


Bruce Labitt wrote:
> Still shagging down a NIC (RTL8169) problem on MythTV. 
> (Reference: 10/16/2009 post, author: Bruce Labitt, Subject: Myth / 
> network question )
>
> My eth1 network connection just stops.  No errors are logged save for 
> link down in message log. 
>
> Problem has persisted through a OS upgrade to Karmic.  I have found 
> reference to 8169 problems at 
> http://ubuntuforums.org/showthread.php?t=538448 but still have no good 
> leads. 
>
> Here is the eth1 message log:
>
> $ dmesg |grep eth1
> [    1.432495] 0000:00:19.0: eth1: (PCI Express:2.5GB/s:Width x1) 
> 00:19:d1:55:47:1a
> [    1.432498] 0000:00:19.0: eth1: Intel(R) PRO/1000 Network Connection
> [    1.432520] 0000:00:19.0: eth1: MAC: 6, PHY: 6, PBA No: ffffff-0ff
> [   13.567380] udev: renamed network interface eth0 to eth1
> [   13.604608] udev: renamed network interface eth1_rename to eth0
> [   15.068741] r8169: eth1: link up
> [   26.023832] eth1: no IPv6 routers present
> [35650.244002] r8169: eth1: link down
> [48370.150267] r8169: eth1: link down
> [48370.150883] ADDRCONF(NETDEV_UP): eth1: link is not ready
> [48382.410469] r8169: eth1: link up
> [48382.411026] ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
> [48393.090009] eth1: no IPv6 routers present
> [61492.984175] r8169: eth1: link down
> [89164.241504] r8169: eth1: link down
> [89164.241934] ADDRCONF(NETDEV_UP): eth1: link is not ready
> [89176.330847] r8169: eth1: link up
> [89176.331396] ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
> [89186.630009] eth1: no IPv6 routers present
> [100142.738728] r8169: eth1: link down
> [126610.257311] r8169: eth1: link down
> [126610.257737] ADDRCONF(NETDEV_UP): eth1: link is not ready
> [126622.743870] r8169: eth1: link up
> [126622.744424] ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
> [126633.190010] eth1: no IPv6 routers present
> [138946.838170] r8169: eth1: link down
> [186290.465991] r8169: eth1: link up
> [187966.299180] r8169: eth1: link up
> [187976.931257] eth1: no IPv6 routers present
> [190045.857011] r8169: eth1: link down
> [190048.613883] r8169: eth1: link up
>
> For those wondering, yes there is a rename of the eth0, eth1 ports.  It 
> is lousy, I know.  Is there a cleaner way to do this, perhaps by MAC so 
> they are assigned properly at boot?
>
> The Realtek 8169 is connected to a gbit switch which in turn is 
> connected to two HDHomerun digital tuners.  I have a DHCP server running 
> on eth1 to give the tuners IPs.
>
> ethtool reports:
> $ sudo ethtool eth1
> [sudo] password for bruce:
> Settings for eth1:
>     Supported ports: [ TP MII ]
>     Supported link modes:   10baseT/Half 10baseT/Full
>                             100baseT/Half 100baseT/Full
>                             1000baseT/Half 1000baseT/Full
>     Supports auto-negotiation: Yes
>     Advertised link modes:  10baseT/Half 10baseT/Full
>                             100baseT/Half 100baseT/Full
>                             1000baseT/Half 1000baseT/Full
>     Advertised auto-negotiation: Yes
>     Speed: 1000Mb/s
>     Duplex: Full
>     Port: MII
>     PHYAD: 0
>     Transceiver: internal
>     Auto-negotiation: on
>     Supports Wake-on: pumbg
>     Wake-on: pumbg
>     Current message level: 0x00000033 (51)
>     Link detected: yes
>
> ifconfig reports:
> $ ifconfig eth1
> eth1      Link encap:Ethernet  HWaddr 00:0e:2e:f0:39:2f 
>           inet addr:192.168.3.1  Bcast:192.168.3.255  Mask:255.255.255.0
>           inet6 addr: fe80::20e:2eff:fef0:392f/64 Scope:Link
>           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
>           RX packets:25609063 errors:0 dropped:0 overruns:0 frame:0
>           TX packets:1513 errors:0 dropped:0 overruns:0 carrier:0
>           collisions:0 txqueuelen:1000
>           RX bytes:35082899041 (35.0 GB)  TX bytes:98189 (98.1 KB)
>           Interrupt:22 Base address:0xe800
>
> ifconfig has not reported any errors when the link (switch link light or 
> ethtool link detect) is down.
>
> I can restart the network, with no problem by sudo 
> /etc/init.d/networking restart  Link light goes on, all is good, for a 
> while.
>
> There are no problems with the eth0 interface, which is an Intel NIC 
> built in to the mobo.  Reliable as all get out.
>
> Needless to say, having the NIC do down at random times ( 1 week, 3.4 
> hours, 34 minutes ) does not make for a reliable media PC.
>
> So, unless anyone has some other things to try, pause, ... any one have 
> a local source for an Intel gigE PCI NIC?
> Can I borrow someones?  Any other ideas?
>
> -Bruce
>
>
> _______________________________________________
> gnhlug-discuss mailing list
> gnhlug-discuss at mail.gnhlug.org
> http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss/
>
>
>