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

Marvell 88E80856 switching from static to dhcp configuration all on its own



Hi All,

I'm having a strange problem with a Marvell 88E8056 - 10/100/1000 Controller on a Biostar TA 770 A2+ motherboard. This is an Etch AMD64 install, but I have added the 2.6.25-amd64 kernel as I could not get the Marvell controller to work at all with the 2.6.18 kernel. Other than that this is pretty much a minimal install. No gui, just a basic system running dns, dhcp3, mysql 5.0, postgresql 8.1and Apache2 in my test lab. This system has no contact with the internet other than as a dns forwarder.

I have two NICs. The onboard Marvell and a 3Com 3c905b. The 3Com handles dhcp and dns requests. Both are configured for statically configured IP addresses in /etc/network/interfaces. However, the Marvell will, after some unknown amount of time--less than 12 hours--drop its static IP address and request a dhcp address from the 3Com adapter. If I do an ifconfig -a it shows the dhcp address, and if I run nmap against the network--how I found this originally as there was an unknown IP address showing up and a known good one missing--the dhcp address shows up. However, if I look at /etc/network/interfaces the Marvell interface is still set to static, and if I restart networking ( /etc/init.d/networking restart ) the Marvell NIC picks up its static IP address again.

Below is the output from lscpi -vv concerning the NIC in question.

03:00.0 Ethernet controller: Marvell Technology Group Ltd. Unknown device 4364 (rev 13)
       Subsystem: Biostar Microtech Int'l Corp Unknown device 2700
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR-
       Latency: 0, Cache Line Size: 32 bytes
       Interrupt: pin A routed to IRQ 1276
       Region 0: Memory at fddfc000 (64-bit, non-prefetchable) [size=16K]
       Region 2: I/O ports at de00 [size=256]
       [virtual] Expansion ROM at fdc00000 [disabled] [size=128K]
       Capabilities: [48] Power Management version 3
Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold+)
               Status: D0 PME-Enable- DSel=0 DScale=1 PME-
       Capabilities: [50] Vital Product Data
Capabilities: [5c] Message Signalled Interrupts: Mask- 64bit+ Queue=0/0 Enable+
               Address: 00000000fee0300c  Data: 4191
       Capabilities: [e0] Express Legacy Endpoint IRQ 0
Device: Supported: MaxPayload 128 bytes, PhantFunc 0, ExtTag-
               Device: Latency L0s unlimited, L1 unlimited
               Device: AtnBtn- AtnInd- PwrInd-
               Device: Errors: Correctable- Non-Fatal- Fatal- Unsupported-
               Device: RlxdOrd- ExtTag- PhantFunc- AuxPwr- NoSnoop-
               Device: MaxPayload 128 bytes, MaxReadReq 512 bytes
               Link: Supported Speed 2.5Gb/s, Width x1, ASPM L0s L1, Port 3
               Link: Latency L0s <256ns, L1 unlimited
               Link: ASPM Disabled RCB 128 bytes CommClk+ ExtSynch-
               Link: Speed 2.5Gb/s, Width x1
       Capabilities: [100] Advanced Error Reporting

Hmmm.... It's been 45 minutes since I restarted networking and the Marvell has already dropped its static IP address and picked up a dhcp address. Below is the relevant information from syslog after restarting networking.
Jul 17 09:45:56 lab kernel: [81245.173294] eth0: no IPv6 routers present
Jul 17 09:46:01 lab kernel: [81249.915550] eth1: no IPv6 routers present
Jul 17 09:46:26 lab dhcpd: receive_packet failed on eth0: Network is down
Jul 17 09:46:26 lab kernel: [81275.403919] sky2 eth1: disabling interface
Jul 17 09:46:26 lab dhclient: receive_packet failed on eth1: Network is down
Jul 17 09:46:26 lab kernel: [81275.414464] eth0:  setting full-duplex.
Jul 17 09:46:26 lab kernel: [81275.449493] sky2 eth1: enabling interface
Jul 17 09:46:26 lab kernel: [81275.451945] ADDRCONF(NETDEV_UP): eth1: link is not ready Jul 17 09:46:29 lab kernel: [81278.425075] sky2 eth1: Link is up at 1000 Mbps, full duplex, flow control both Jul 17 09:46:29 lab kernel: [81278.427530] ADDRCONF(NETDEV_CHANGE): eth1: link becomes ready
Jul 17 09:46:36 lab kernel: [81285.424608] eth0: no IPv6 routers present
Jul 17 09:46:39 lab kernel: [81288.428601] eth1: no IPv6 routers present

Jul 17 09:55:19 lab dhclient: DHCPREQUEST on eth1 to 192.168.1.28 port 67
Jul 17 09:55:33 lab last message repeated 3 times

Jul 17 09:55:41 lab dhclient: DHCPREQUEST on eth1 to 192.168.1.28 port 67
Jul 17 09:56:18 lab last message repeated 3 times
Jul 17 09:57:38 lab last message repeated 4 times
Jul 17 09:58:39 lab last message repeated 4 times
Jul 17 09:59:29 lab last message repeated 3 times
Jul 17 10:00:28 lab last message repeated 4 times

This repeats a few more times and then:

Jul 17 10:04:39 lab dhcpd: Wrote 0 deleted host decls to leases file.
Jul 17 10:04:39 lab dhcpd: Wrote 0 new dynamic host decls to leases file.
Jul 17 10:04:39 lab dhcpd: Wrote 7 leases to leases file.

Jul 17 10:06:57 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:06:57 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:07:06 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:07:06 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:07:06 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:07:14 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:07:14 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:07:14 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:07:27 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:07:27 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:07:27 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:07:42 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:07:42 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:07:42 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:07:50 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:07:50 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:07:50 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:08:08 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:08:08 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:08:08 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:08:15 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:08:15 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:08:15 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:08:29 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:08:29 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:08:29 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:08:37 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:08:37 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:08:37 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:08:47 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:08:47 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:08:47 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:09:04 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:09:04 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:09:04 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:09:25 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:09:25 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:09:25 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:09:41 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:09:41 lab dhcpd: DHCPREQUEST for 192.168.1.40 from 00:e0:4d:7b:82:df via eth0 Jul 17 10:09:41 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0 Jul 17 10:09:57 lab dhclient: DHCPDISCOVER on eth1 to 255.255.255.255 port 67 interval 5
Jul 17 10:09:57 lab dhcpd: DHCPDISCOVER from 00:e0:4d:7b:82:df via eth0
Jul 17 10:09:57 lab dhcpd: DHCPOFFER on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:09:57 lab dhclient: DHCPOFFER from 192.168.1.28
Jul 17 10:09:57 lab dhclient: DHCPREQUEST on eth1 to 255.255.255.255 port 67
Jul 17 10:09:57 lab dhcpd: DHCPREQUEST for 192.168.1.40 (192.168.1.28) from 00:e0:4d:7b:82:df via eth0 Jul 17 10:09:57 lab dhcpd: DHCPACK on 192.168.1.40 to 00:e0:4d:7b:82:df via eth0
Jul 17 10:09:57 lab dhclient: DHCPACK from 192.168.1.28
Jul 17 10:09:57 lab dhclient: bound to 192.168.1.40 -- renewal in 707 seconds.

I'm assuming this is a bug in the sky2 module, but don't know enough about things in this area to do more than assume. BTW, I have a problem with Biostar A770 A2+ and an onboard RealTek 8111/8168B chipset and same kernel only this is a desktop Sid install. It's even odder. If I leave the NIC (once again 2 nics in the system) in the 'up" state for more than 10 or 15 minutes I lose the ability to access the internet with Firefox. I can access any site on my local http server, ping out on the internet, resolve dns requests, use ftp, ssh, etc... but Firefox hangs "connecting to" whatever Internet web site I'm trying to reach. As soon as I ifdown the ethx connection my internet connectivity in Firefox comes back. The Firefox http internet connection will be made and work as soon as the offending NIC is taken offline. Once again, if I restart networking things will work fine for 10 to 15 minutes and then the ability to reach any internet web site is gone. If I leave the onboard NIC disabled, no problems whatsover.
Reply to: