January 23, 201016 yr I have a server I've been using over 2 years and a server I built 2 days ago. Both appear to boot fine to login prompt. One has a static IP one uses DHCP. Nothing has changed in either's config as far as I know. I can no longer ping either server. I can't reach the http pages for either. emhttp process appears to be running on both. I cannot access the servers by IP address (for the static one, the DHCP doesn't appear to be getting an IP). I have reset the router and verified all network connections wires. What in the heck could cause such a thing? The only thing that changes was I tried to add more drives to the second system. They both appear to boot to tty login fine ... The older system uses the name 'galaxy' the new system still uses 'tower'. I've attached several syslogs as follows: syslog_galaxy_static_ip - system in use for almost 2 years. has a static IP. syslog_noip_dhcp - system in use 2 days. 'tower'. all disks attached syslog_noip_dhcp_nodisks - same as above. all extra peripherals disconnected. Do these indicate something? What in the heck would do such a thing? Thanks so much for any help you can offer. syslog_galaxy_static_ip.txt syslog_noip_dhcp.txt syslog_noip_dhcp_nodisks.txt
January 23, 201016 yr Author Additional info - galaxy ( server 1 ) worked 20 minutes ago. Nothing changes as far a sI know. It is unable to ping the gateway (192.168.1.1) destination host unreachable. Wires are sound. NIC tested in different system, works fine. eth0 Link encap:Ethernet HWaddr 00:1b:21:0e:e6:30 inet addr:192.168.1.22 Bcast:192.168.1.255 Mask:255.255.255.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0 TX packets:73 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:0 (0.0 B) TX bytes:8090 (7.9 KiB)
January 23, 201016 yr I didn't check your logs - but I had the same in the past and it was caused by the fact, that both unraid servers used the SAME MAC address (I copied config...). Thus you get problems with your switch (layer 2). If this applies in your case, just delete the MAC address or use differnt ones. regards, Guzzi
January 23, 201016 yr Author well only one of the servers is on right now but where do I set the mac address? I'll research this. edit: I don't have hardware address set in network.cfg. I changed the static IP on the server and unplugged modem / router for a few minutes to see if anything diff happens.
January 23, 201016 yr I have a server I've been using over 2 years and a server I built 2 days ago. Both appear to boot fine to login prompt. One has a static IP one uses DHCP. Nothing has changed in either's config as far as I know. I can no longer ping either server. I can't reach the http pages for either. emhttp process appears to be running on both. I cannot access the servers by IP address (for the static one, the DHCP doesn't appear to be getting an IP). I have reset the router and verified all network connections wires. What in the heck could cause such a thing? Galaxy seems to be OK: Jan 23 03:45:07 galaxy kernel: e1000: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX Jan 23 03:45:07 galaxy ifplugd(eth0)[1193]: Executing '/etc/ifplugd/ifplugd.action eth0 up'. Jan 23 03:45:07 galaxy logger: /etc/rc.d/rc.inet1: /sbin/ifconfig eth0 192.168.1.22 broadcast 192.168.1.255 netmask 255.255.255.0 Jan 23 03:45:07 galaxy logger: /etc/rc.d/rc.inet1: /sbin/route add default gw 192.168.1.1 metric 1 Tower is having a problem getting a response from DHCP server: Jan 23 06:36:25 Tower dhcpcd[1220]: broadcasting DHCP_DISCOVER Jan 23 06:36:55 Tower dhcpcd[1220]: timed out waiting for a valid DHCP server response What device on your network is playing a DHCP server? Did you do a fresh install on the new server, or did you copy stuff from the flash key of the old one? Try this: On the console of Tower execute this command: ifconfig eth0 192.168.1.123 netmask 255.255.255.0 up ...then point a web browser to address 192.168.1.123 You should be able to see your new server. Then you can manage its settings from there. Purko
January 23, 201016 yr Author A D-Link router is the DHCP server. It is working for all other devices in the house minus unraid. I'll try your suggestion regarding Galaxy.
January 23, 201016 yr Author I tried your suggestion. It seems pointless and neither system can ping the router I'm afraid. No change.
January 23, 201016 yr I tried your suggestion. It seems pointless and neither system can ping the router I'm afraid. No change. Each unRAID server has its own config/network.cfg In each network.cfg is a line that is used to set the MAC address of the server. This is a field that is shown on the unRAID settings screen, but it is not an enter-able field. If you accidentally end up with the MAC address of another PC on your LAN, because you initially used the flash drive to test booting up with the other PC, it will grab the MAC address of the other network controller. When that happens, both will try to respond to the same packets and lots of things will stop working. The line in the config/network.cfg file that sets the HWADDR may be deleted. You can do that easily by running the following command at a linux prompt. sed -i "/HWADDR/d" /boot/config/network.cfg I'd do that on both your servers, then reboot both of them AND your router. (You need to reboot the router too, so it forgets all it knows about the IP address/MAC address associations of the machines involved.) Don't worry about the HWADDR line, the network chipset on your server will use its native one. Also, just to be absolutely certain, since you are using static IP addresses, make sure the IP address of each of your servers is unique from everything else on your LAN. Also, if you change any IP address and have edited local "hosts" files, the files will need to be updated with the new IP address. Joe L.
January 23, 201016 yr Author gIn each network.cfg is a line that is used to set the MAC address of the server. This is a field that is shown on the unRAID settings screen, but it is not an enter-able field. If you accidentally end up with the MAC address of another PC on your LAN, because you initially used the flash drive to test booting up with the other PC, it will grab the MAC address of the other network controller. Hi Joe, I never had that line in my network config. The network config file only has USE_DCHP=yes at this point. previous it has a static ip and gateway address. The machine is not getting an IP from a router and cannot ping the router. This went from working to 20 minutes later not working. The router is still giving IPs to 3 Win 7 machines (2 wireless). This happened on both systems at approximately the same time. I've tried different PCI slots for both NICs, tested the NICs in the Win7 machines, changes out 4 different ethernet wires, and verified the ethernet wires are soundly connected to the Linux boxes. (the lights on the port light, and the box beeps when the wire is disconnected) I'm drawing a blank here.
January 23, 201016 yr I'll try your suggestion regarding Galaxy. ... I tried your suggestion. It seems pointless My suggestion was not regarding Galaxy! It was for your new server, to get it up so it can be accessile by the web interface. And also, I'll ask again... Did you do a fresh install on the new server, or did you copy stuff from the flash key of the old one?
January 23, 201016 yr Author My suggestion was not regarding Galaxy! It was for your new server, to get it up so it can be accessile by the web interface. And also, I'll ask again... Did you do a fresh install on the new server, or did you copy stuff from the flash key of the old one? I'm sorry, I meant to say Tower. I set the IP as your specified on Tower and it does not work still. It is now in a situation like galaxy. It has an IP but cannot ping the gateway or be accessible via the web / telnet. (emhttp process is running) Tower was set up as a completely fresh install. I downloaded Unraid 4.5, installed on a brand new flash and installed 2 drives. At this time it was accessible during the web. I added several new drives and started her up ... no web access and the syslog you see now. So, I unplugged all the drives and tried again (still no web access). Something very odd has happened. The ubuntu box on the router is working fine with DHCP as are my 3 windows machines.
January 23, 201016 yr The ubuntu box on the router is working fine with DHCP as are my 3 windows machines. What IP addresses do the 3 windows machines show? Can you please post here the network.cfg files from the flash keys of the two unRAID servers?
January 23, 201016 yr Author The ubuntu box on the router is working fine with DHCP as are my 3 windows machines. What IP addresses do the 3 windows machines show? Can you please post here the network.cfg files from the flash keys of the two unRAID servers? sam-PC - 192.168.1.106 Aquilas - 192.168.1.114 TOWER Network.Cfg: # Generated network settings USE_DHCP=yes IPADDR= NETMASK= GATEWAY= GALAXY Network.Cfg: ( new version ) # Generated network settings USE_DHCP=yes Galaxy Network.Cfg (old version # Generated network settings USE_DHCP=no IPADDR=192.168.1.22 NETMASK=255.255.255.0 GATEWAY=192.168.1.1
January 23, 201016 yr The ubuntu box on the router is working fine with DHCP as are my 3 windows machines. What IP addresses do the 3 windows machines show? Can you please post here the network.cfg files from the flash keys of the two unRAID servers? sam-PC - 192.168.1.106 Aquilas - 192.168.1.114 TOWER Network.Cfg: # Generated network settings USE_DHCP=yes IPADDR= NETMASK= GATEWAY= GALAXY Network.Cfg: ( new version ) # Generated network settings USE_DHCP=no Galaxy Network.Cfg (old version # Generated network settings USE_DHCP=no IPADDR=192.168.1.22 NETMASK=255.255.255.0 GATEWAY=192.168.1.1 Glaxy was just fine before. What do you expect of the "new version"? It has no static IP, and use DHCP=no. Power down ALL devices on your network, including the DHCP server. Then turn on only the DHCP server, and then TOwer. DOes TOwer get a IP address? (on the console of Tower type ifconfig) Also, try this on the console of TOwer: cat /var/log/syslog | grep DHCP Or: cat /var/log/syslog | grep eth0
January 23, 201016 yr FWIW - When you power down your unraid units, pull the power cord out of the PSU then press the power button as if you were restarting them. Wait a minute before reconnection the power cords. Strange as it may sound, I have had a motherboard hold residual power, not clearing a network fault, till I discharged the Power Supply completely. May make a difference, may not. But no harm in trying. My two cents. Rob
January 23, 201016 yr Author Power down ALL devices on your network, including the DHCP server. Then turn on only the DHCP server, and then TOwer. DOes TOwer get a IP address? (on the console of Tower type ifconfig) Also, try this on the console of TOwer: cat /var/log/syslog | grep DHCP Or: cat /var/log/syslog | grep eth0 Power down everything, DHCP discovery times out on Tower. I put in a new NIC. Does not work. Assigned a static IP to it like you said. Hey, it works, awesome! Close up the box. Turn it on ... the box no longer has network access (has a static IP, but can't ping 192.168.1.1 gate way or be reached on the network... what the hell? I swap the cord back to the on-board NIC, assigned static IP. magic! it works again. Close it up, turn it back on .... no connectivity. I'm going to shoot myself in the head.
March 11, 201016 yr Hopefully you did not shoot yourself?!? I'm obviously too late to stop you! This may be off the wall, but try saving all settings of the router to a local file, then resetting the router. It sounds to me like a router issue.
Archived
This topic is now archived and is closed to further replies.