SG1100 dead NICs
-
@kibade said in SG1100 dead NICs:
(Lighting damage is not covered under the warranty)
That why there is ( should be !!?) a second assurance that covers for lighting / fire / floods / theft / etc. Your 'home' assurance or whatever it's being called in your country.
-
@stephenw10 Thanks for the advice. I run the command in a shell
etherswitchcfg: Can't open control file: /dev/etherswitch0: No such file or directory
If I list the dev folder I can see etherswitchX is missing :
I assume it doesn't see it on startup.
-
@gertjan They could, but I suspect it would increase the site's business insurance the more claims they make. I might invest in an RJ45 lighting Suppressor for the WAN. The ISP is a wireless provider. I suspect the transceiver is the entry point, but no way to proved that. If I get time I might inspect the ISP install, focusing on grounding. However, that's outside my domain.
-
@kibade said in SG1100 dead NICs:
I assume it doesn't see it on startup.
Reboot and check the console. You should see in boot log:
mvneta0: <NETA controller> mem 0x30000-0x33fff irq 14 on simplebus1 mvneta0: version is 10 mvneta0: Ethernet address: f0:ad:4e:08:6b:df mdio0: <MDIO> on mvneta0 e6000sw0: <Marvell 88E6341> on mdio0 e6000sw0: multi-chip addressing mode (0x1) e6000sw0: CPU port at 0 e6000sw0: fixed port at 0 e6000sw0: PHY at port 1 miibus0: <MII bus> on e6000sw0 e1000phy0: <Marvell 88E1000 Gigabit PHY> PHY 17 on miibus0 e1000phy0: none, 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-master, 1000baseT-FDX, 1000baseT-FDX-master, auto e6000sw0: PHY at port 2 miibus1: <MII bus> on e6000sw0 e1000phy1: <Marvell 88E1000 Gigabit PHY> PHY 18 on miibus1 e1000phy1: none, 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-master, 1000baseT-FDX, 1000baseT-FDX-master, auto e6000sw0: PHY at port 3 miibus2: <MII bus> on e6000sw0 e1000phy2: <Marvell 88E1000 Gigabit PHY> PHY 19 on miibus2 e1000phy2: none, 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-master, 1000baseT-FDX, 1000baseT-FDX-master, auto e6000sw0: switch is ready. etherswitch0: <Switch controller> on e6000sw0Steve
-
Is see this in dmesg:-
mvneta0: <NETA controller> mem 0x30000-0x33fff irq 14 on simplebus1 mvneta0: version is 10 mvneta0: Ethernet address: f0:ad:4e:19:28:29 mdio0: <MDIO> on mvneta0 e6000sw: readreg timeout e6000sw0: Unrecognized device, id 0xfff0. e6000sw: readreg timeout e6000sw0: Unrecognized device, id 0xfff0.And this in syslog (note the date is wrong)
Aug 4 16:57:34 VCF-fw kernel: Warning: no time-of-day clock registered, system time will not be set accurately Aug 4 16:57:34 VCF-fw php[2966]: rc.bootup: The command '/sbin/etherswitchcfg -f /dev/etherswitch0 port0 forwarding' returned exit code '69', the output was 'etherswitchcfg: Can't open control file: /dev/etherswitch0: No such file or directory' Aug 4 16:57:34 VCF-fw php[2966]: rc.bootup: The command '/sbin/etherswitchcfg -f /dev/etherswitch0 port1 forwarding' returned exit code '69', the output was 'etherswitchcfg: Can't open control file: /dev/etherswitch0: No such file or directory' Aug 4 16:57:34 VCF-fw php[2966]: rc.bootup: The command '/sbin/etherswitchcfg -f /dev/etherswitch0 port2 forwarding' returned exit code '69', the output was 'etherswitchcfg: Can't open control file: /dev/etherswitch0: No such file or directory' Aug 4 16:57:34 VCF-fw php[2966]: rc.bootup: The command '/sbin/etherswitchcfg -f /dev/etherswitch0 port3 forwarding' returned exit code '69', the output was 'etherswitchcfg: Can't open control file: /dev/etherswitch0: No such file or directory' -
Mmm, looks like the switch IC is toast then. Nothing much you can do from there unfortunately.
Add USB NICs? I would not recommend that in any other situation.
Steve
-
@stephenw10 I thought as much. Your help has been awesome, thanks. I did look at using a spare USB to NIC. The OS saw it OK, but that's as far as I went, looked too quirky. How hard would it be to swap out the IC? I'm pretty good at soldering?
-
It's all tiny surface mount stuff so....hard! Also it's impossible to say what else might be damaged. The driver is not able to see the switch on the NIC MDIO lines. But that could still be the NIC. Or something on the MDIO bus. Or all of that.
Steve
-
@stephenw10 If it smd and not BGA I prob could do it even though It's been a while since I was an official Electronics Engineer (before moving into IT) I think you are right, even if the switches IC is damaged it is most likely not the only item. I have a replacement SG1100 in front of me and now an APC RJ45 surge suppressor. Just need to chat with the site electrician to make sure the ground connections are correctly in place. It's unlikely, but I want to be prepared for a second time.
I've discovered, same building and same ISP, diff company lost their WAN-connected router, switch, and printer on the same day! Seems that somehow some of the lighting power coupled into the WAN from the ISP and then into both company's main routers. Odd the ISP was still running. I asked around and found another customer of the same ISP (car mechanic) whose router and their ISP transceiver die that same day (The lighting was crazy that day). I will make it a standard requirement from now on when using that ISP we use a good RJ45 surge suppressor.
Your support has been amazing. Is there somewhere you would like me to put some positive feedback? -
You can open a ticket or just email support@netgate.com. All feedback is appreciated.
-
@stephenw10 Looks like bandartogel is trying to use your forum to advertise!
