March 2, 200917 yr We lost power in the middle of the night last night - about 3-4 times over 5-10 minutes. With UPS in place, I was hopeful that all would be well. This morning, I was not able to see my unRAID server on the network. Upon examining, I found that unRAID had not shutdown, and was responsive to the console. (Note the router and switches are not on the UPS). I used WeeboTech's powerdown script which cleanly shutdown the server and booted back up cleanly. (Thanks Weebo!) All is well. No parity check. Is there was a command I could have entered at the console that would have brought the array back on the network without a reboot? Here is my syslog starting with the first power failure. (The first THREE reported power failures were completely unnoticed by anyone in my house. We were in the middle of watchin a movie off the unRAID server!.) Mar 1 21:20:46 Tower apcupsd[2837]: Power failure. Mar 1 21:20:48 Tower apcupsd[2837]: Power is back. UPS running on mains. Mar 1 21:42:12 Tower apcupsd[2837]: Power failure. Mar 1 21:42:13 Tower apcupsd[2837]: Power is back. UPS running on mains. Mar 1 21:51:12 Tower dhcpcd[2234]: sending DHCP_REQUEST for 192.168.216.41 to 192.168.216.199 Mar 1 21:51:12 Tower dhcpcd[2234]: dhcpIPaddrLeaseTime=43201 in DHCP server response. Mar 1 21:51:12 Tower dhcpcd[2234]: DHCP_ACK received from (192.168.216.199) Mar 1 21:59:18 Tower apcupsd[2837]: Power failure. Mar 1 21:59:19 Tower apcupsd[2837]: Power is back. UPS running on mains. Mar 2 03:51:12 Tower dhcpcd[2234]: sending DHCP_REQUEST for 192.168.216.41 to 192.168.216.199 Mar 2 03:51:12 Tower dhcpcd[2234]: dhcpIPaddrLeaseTime=86400 in DHCP server response. Mar 2 03:51:12 Tower dhcpcd[2234]: DHCP_ACK received from (192.168.216.199) Mar 2 04:46:54 Tower apcupsd[2837]: Power failure. Mar 2 04:46:55 Tower kernel: e1000: eth0: e1000_watchdog_task: NIC Link is Down Mar 2 04:46:55 Tower ifplugd(eth0)[2345]: Link beat lost. Mar 2 04:47:00 Tower apcupsd[2837]: Running on UPS batteries. Mar 2 04:47:06 Tower ifplugd(eth0)[2345]: Executing '/etc/ifplugd/ifplugd.action eth0 down'. Mar 2 04:47:06 Tower logger: /etc/rc.d/rc.inet1: /sbin/dhcpcd -k -d eth0 Mar 2 04:47:06 Tower dhcpcd[2234]: sending DHCP_RELEASE for 192.168.216.41 to 192.168.216.199 Mar 2 04:47:07 Tower dhcpcd[2234]: terminating on signal 1 Mar 2 04:47:07 Tower ifplugd(eth0)[2345]: Program executed successfully. Mar 2 04:47:15 Tower apcupsd[2837]: Mains returned. No longer on UPS batteries. Mar 2 04:47:15 Tower apcupsd[2837]: Power is back. UPS running on mains. Mar 2 04:47:18 Tower kernel: e1000: eth0: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX Mar 2 04:47:18 Tower ifplugd(eth0)[2345]: Link beat detected. Mar 2 04:47:19 Tower apcupsd[2837]: Power failure. Mar 2 04:47:19 Tower ifplugd(eth0)[2345]: Executing '/etc/ifplugd/ifplugd.action eth0 up'. Mar 2 04:47:19 Tower ifplugd(eth0)[2345]: client: Polling for DHCP server on interface eth0: Mar 2 04:47:20 Tower kernel: e1000: eth0: e1000_watchdog_task: NIC Link is Down Mar 2 04:47:21 Tower ifplugd(eth0)[2345]: client: No carrier detected on eth0. Reducing DHCP timeout to 10 seconds. Mar 2 04:47:21 Tower logger: /etc/rc.d/rc.inet1: /sbin/dhcpcd -d -t 10 -h Tower eth0 Mar 2 04:47:21 Tower dhcpcd[25811]: broadcasting DHCP_DISCOVER Mar 2 04:47:25 Tower apcupsd[2837]: Running on UPS batteries. Mar 2 04:47:31 Tower dhcpcd[25811]: timed out waiting for a valid DHCP server response Mar 2 04:47:31 Tower ifplugd(eth0)[2345]: client: dhcpcd: MAC address = 00:1b:fc:62:7d:08 Mar 2 04:47:31 Tower ifplugd(eth0)[2345]: Program executed successfully. Mar 2 04:47:31 Tower ifplugd(eth0)[2345]: Link beat lost. Mar 2 04:47:42 Tower ifplugd(eth0)[2345]: Executing '/etc/ifplugd/ifplugd.action eth0 down'. Mar 2 04:47:42 Tower logger: /etc/rc.d/rc.inet1: /sbin/dhcpcd -k -d eth0 Mar 2 04:47:43 Tower ifplugd(eth0)[2345]: Program executed successfully. Mar 2 04:47:56 Tower apcupsd[2837]: Mains returned. No longer on UPS batteries. Mar 2 04:47:56 Tower apcupsd[2837]: Power is back. UPS running on mains. Mar 2 04:47:58 Tower kernel: e1000: eth0: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX Mar 2 04:47:59 Tower ifplugd(eth0)[2345]: Link beat detected. Mar 2 04:48:00 Tower ifplugd(eth0)[2345]: Executing '/etc/ifplugd/ifplugd.action eth0 up'. Mar 2 04:48:00 Tower ifplugd(eth0)[2345]: client: Polling for DHCP server on interface eth0: Mar 2 04:48:01 Tower logger: /etc/rc.d/rc.inet1: /sbin/dhcpcd -d -t 30 -h Tower eth0 Mar 2 04:48:02 Tower dhcpcd[25874]: broadcasting DHCP_DISCOVER Mar 2 04:48:04 Tower kernel: e1000: eth0: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX Mar 2 04:48:31 Tower dhcpcd[25874]: broadcastAddr option is missing in DHCP server response. Assuming 192.168.216.255 Mar 2 04:48:31 Tower dhcpcd[25874]: dhcpIPaddrLeaseTime=86400 in DHCP server response. Mar 2 04:48:31 Tower dhcpcd[25874]: DHCP_OFFER received from (192.168.216.199) Mar 2 04:48:31 Tower dhcpcd[25874]: broadcasting DHCP_REQUEST for 192.168.216.30 Mar 2 04:48:31 Tower dhcpcd[25874]: timed out waiting for a valid DHCP server response Mar 2 04:48:32 Tower ifplugd(eth0)[2345]: client: dhcpcd: MAC address = 00:1b:fc:62:7d:08 Mar 2 04:48:32 Tower ifplugd(eth0)[2345]: Program executed successfully. Mar 2 04:48:32 Tower ifplugd(eth0)[2345]: Link beat lost. Mar 2 04:48:35 Tower kernel: e1000: eth0: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX Mar 2 04:48:35 Tower ifplugd(eth0)[2345]: Link beat detected. Mar 2 04:51:13 Tower ntpd[2321]: Deleting interface #2 eth0, 192.168.216.41#123, interface stats: received=2875, sent=2901, dropped=0, active_time=2487601 secs Mar 2 07:19:01 Tower logger: Powerdown initiated ...
March 2, 200917 yr Server douldn't get an IP address... I'm assuming your router/switch was the DHCP server and it was down, and it may have been too slow when power was restored. I would get rid of DHCP and assign a static IP to the server.... and put a UPS on the router/switch and modem. Then you can still have Internet access during a power outage.
March 2, 200917 yr Author Thanks for the advice, BubbaQ. I will have to do some "redecorating" to make that happen. Still have the question, though .. Is there was a command I could have entered at the console that would have brought the array back on the network without a reboot? In Windows you can do a "renew". Is there an equivalent in Linux?
March 2, 200917 yr Thanks for the advice, BubbaQ. I will have to do some "redecorating" to make that happen. Still have the question, though .. Is there was a command I could have entered at the console that would have brought the array back on the network without a reboot? In Windows you can do a "renew". Is there an equivalent in Linux? ifconfig [device] release/renew i.e.: ifconfig fxp0 renew not sure if this works with unRAID and I'm not at home to test
Archived
This topic is now archived and is closed to further replies.