August 22, 201510 yr Community Expert I just got a notification from UPS that power had been restored, but I didn't get any notification that it had failed. I know I have gotten power fail notifications before so wonder if this is something that got broken in recent 6.1rc Looking at syslog, it was a very brief event, and I didn't even see any lights dim or anything. Probably time for me to invest in an UPS with AVR. Aug 22 09:17:34 unSERVER dhcpcd[1429]: eth0: xid 0x95612c0a is for hwaddr 00:26:9e:68:b1:46:00:00:00:00:00:00:00:00:00:00 Aug 22 11:37:32 unSERVER apcupsd[11599]: Power failure. Aug 22 11:37:43 unSERVER apcupsd[11599]: Running on UPS batteries. Aug 22 11:37:43 unSERVER apcupsd[11599]: Mains returned. No longer on UPS batteries. Aug 22 11:37:43 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 22 11:37:43 unSERVER sSMTP[29647]: Creating SSL connection to host Aug 22 11:37:43 unSERVER sSMTP[29647]: SSL connection using ECDHE-RSA-AES128-GCM-SHA256 Aug 22 11:37:46 unSERVER sSMTP[29647]: Sent mail for...
August 22, 201510 yr Do you mean an email notification didn't. Come through or a notification in the webgui? If power was lost, wouldn't your internet be down too? If you mean a notification in the webgui, I'll have to look into this on Monday.
August 22, 201510 yr Author Community Expert Do you mean an email notification didn't. Come through or a notification in the webgui? If power was lost, wouldn't your internet be down too? If you mean a notification in the webgui, I'll have to look into this on Monday. No notification in the webgui or email for the power failure, but I got both email and webgui notifications on the power restore. Also, it was not a true power loss, probably just my UPS detecting a power drop or something. I didn't even see any lights flicker or dim when it happened. And all of my network is on UPS anyway so the internet wasn't down. If you look at the syslog snippet I posted you can see the email being sent for the power restore but none sent for the power fail.
August 22, 201510 yr Odd. I will try to recreate on Monday. Don't have a UPS here at home to test with. Any chance you can try testing again by unplugging the UPS from the wall and seeing if the notification goes out? Want to know if it was a fluke or is reproducible consistently.
August 22, 201510 yr Community Expert Reading this I remembered the same thing happened to me a few days ago with v6.0.1, server is still on so here is the relevant info from the log, can upload complete log if necessary. Aug 15 03:40:01 Tower5 logger: mover finished Aug 15 12:26:54 Tower5 apcupsd[21488]: Power failure. Aug 15 12:26:56 Tower5 kernel: usb usb4-port5: disabled by hub (EMI?), re-enabling... Aug 15 12:26:56 Tower5 kernel: usb 4-5: USB disconnect, device number 2 Aug 15 12:26:56 Tower5 kernel: usb 4-5: new full-speed USB device number 3 using ohci-pci Aug 15 12:26:56 Tower5 kernel: hid-generic 0003:051D:0002.0002: hiddev0,hidraw0: USB HID v1.00 Device [American Power Conversion Back-UPS BR 900GI FW:879.L1d.I USB FW:L1d ] on usb-0000:00:12.0-5/input0 Aug 15 12:27:02 Tower5 apcupsd[21488]: Communications with UPS restored. Aug 15 12:27:02 Tower5 apcupsd[21488]: Running on UPS batteries. Aug 15 12:27:02 Tower5 apcupsd[21488]: Mains returned. No longer on UPS batteries. Aug 15 12:27:02 Tower5 apcupsd[21488]: Power is back. UPS running on mains. Aug 15 12:27:03 Tower5 sSMTP[26838]: Creating SSL connection to host Aug 15 12:27:03 Tower5 sSMTP[26838]: SSL connection using ECDHE-RSA-AES128-GCM-SHA256 Aug 15 12:27:06 Tower5 sSMTP[26838]: Sent mail for [email protected] (221 2.0.0 closing connection lz10sm12435990wjb.48 - gsmtp) uid=0 username=root outbytes=674 Aug 15 12:36:15 Tower5 kernel: mdcmd (34): spindown 2 The mail I got was that communication with ups was restored. My router is also on the UPS so never lost internet.
August 22, 201510 yr Author Community Expert Odd. I will try to recreate on Monday. Don't have a UPS here at home to test with. Any chance you can try testing again by unplugging the UPS from the wall and seeing if the notification goes out? Want to know if it was a fluke or is reproducible consistently. OK, I tried the test, and also searched my syslog to see what else I could see related to this. Aug 16 14:46:53 unSERVER kernel: mdcmd (31): spindown 0 Aug 16 18:16:37 unSERVER apcupsd[11599]: Power failure. Aug 16 18:16:48 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 16 18:16:48 unSERVER apcupsd[11599]: Power failure. Aug 16 18:16:48 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 16 18:25:09 unSERVER in.telnetd[18097]: connect from 10.0.0.2 (10.0.0.2) No notifications on Power failure or Power is back, and no other evidence that there was an actual power outage. Aug 19 17:24:34 unSERVER kernel: mdcmd (92): spindown 1 Aug 19 17:34:59 unSERVER apcupsd[11599]: Power failure. Aug 19 17:35:01 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 19 17:49:14 unSERVER kernel: mdcmd (93): spindown 1 No notifications on Power failure or Power is back, and no other evidence that there was an actual power outage. Aug 19 18:16:04 unSERVER dhcpcd[1429]: eth0: ARP announcing 10.0.0.3 (2 of 2) Aug 19 18:31:10 unSERVER apcupsd[11599]: Power failure. Aug 19 18:31:12 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 19 18:33:43 unSERVER apcupsd[11599]: Power failure. Aug 19 18:33:45 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 19 18:34:31 unSERVER apcupsd[11599]: Power failure. Aug 19 18:34:33 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 19 21:10:05 unSERVER dhcpcd[1429]: eth0: xid 0xa20c92e7 is for hwaddr 00:18:dd:05:14:8e:00:00:00:00:00:00:00:00:00:00 No notifications on Power failure or Power is back, and no other evidence that there was an actual power outage. Aug 21 17:34:23 unSERVER login[14414]: ROOT LOGIN on '/dev/pts/1' from 'Yoga' Aug 21 19:09:09 unSERVER apcupsd[11599]: Power failure. Aug 21 19:09:11 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 21 19:11:24 unSERVER apcupsd[11599]: Power failure. Aug 21 19:11:34 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 21 19:11:34 unSERVER apcupsd[11599]: Power failure. Aug 21 19:11:35 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 21 19:39:33 unSERVER kernel: mdcmd (29): spindown 0 No notifications on Power failure or Power is back, and no other evidence that there was an actual power outage. Aug 22 09:17:34 unSERVER dhcpcd[1429]: eth0: xid 0x95612c0a is for hwaddr 00:26:9e:68:b1:46:00:00:00:00:00:00:00:00:00:00 Aug 22 11:37:32 unSERVER apcupsd[11599]: Power failure. Aug 22 11:37:43 unSERVER apcupsd[11599]: Running on UPS batteries. Aug 22 11:37:43 unSERVER apcupsd[11599]: Mains returned. No longer on UPS batteries. Aug 22 11:37:43 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 22 11:37:43 unSERVER sSMTP[29647]: Creating SSL connection to host Aug 22 11:37:43 unSERVER sSMTP[29647]: SSL connection using ECDHE-RSA-AES128-GCM-SHA256 Aug 22 11:37:46 unSERVER sSMTP[29647]: Sent mail for... This is the event that I reported earlier. No notification on Power failure, but got notifications this time on Power is back. No other evidence that there was an actual power outage. Aug 22 13:02:08 unSERVER kernel: mdcmd (40): spindown 4 Aug 22 13:16:06 unSERVER emhttp: /usr/local/emhttp/webGui/scripts/tail_log syslog 2>&1 Aug 22 13:16:43 unSERVER apcupsd[11599]: Power failure. Aug 22 13:16:49 unSERVER apcupsd[11599]: Running on UPS batteries. Aug 22 13:16:49 unSERVER sSMTP[5337]: Creating SSL connection to host Aug 22 13:16:49 unSERVER sSMTP[5337]: SSL connection using ECDHE-RSA-AES128-GCM-SHA256 Aug 22 13:16:51 unSERVER sSMTP[5337]: Sent mail for [email protected] (221 2.0.0 closing connection y66sm11382543ywe.11 - gsmtp) uid=0 username=root outbytes=702 Aug 22 13:17:39 unSERVER apcupsd[11599]: Mains returned. No longer on UPS batteries. Aug 22 13:17:39 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 22 13:17:40 unSERVER sSMTP[6235]: Creating SSL connection to host Aug 22 13:17:40 unSERVER sSMTP[6235]: SSL connection using ECDHE-RSA-AES128-GCM-SHA256 Aug 22 13:17:41 unSERVER sSMTP[6235]: Sent mail for... The unplug test. Got notifications for both Power failure and Power is back. There was definitely a power outage to the UPS this time, because I had unplugged it from the wall. Don't know anything about the actual code, but I'm thinking the UPS is probably signalling low voltage and actual power out differently, and unRAID is using this difference to determine whether or not to do a notification, but it is putting Power failure in the syslog regardless of which type of event happened.
August 22, 201510 yr I just got a notification from UPS that power had been restored, but I didn't get any notification that it had failed. I know I have gotten power fail notifications before so wonder if this is something that got broken in recent 6.1rc Looking at syslog, it was a very brief event, and I didn't even see any lights dim or anything. Probably time for me to invest in an UPS with AVR. Aug 22 09:17:34 unSERVER dhcpcd[1429]: eth0: xid 0x95612c0a is for hwaddr 00:26:9e:68:b1:46:00:00:00:00:00:00:00:00:00:00 Aug 22 11:37:32 unSERVER apcupsd[11599]: Power failure. Aug 22 11:37:43 unSERVER apcupsd[11599]: Running on UPS batteries. Aug 22 11:37:43 unSERVER apcupsd[11599]: Mains returned. No longer on UPS batteries. Aug 22 11:37:43 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 22 11:37:43 unSERVER sSMTP[29647]: Creating SSL connection to host Aug 22 11:37:43 unSERVER sSMTP[29647]: SSL connection using ECDHE-RSA-AES128-GCM-SHA256 Aug 22 11:37:46 unSERVER sSMTP[29647]: Sent mail for... I haven't tested this is a while, but I don't believe a power failure event is reported that fast. If I remember correctly, the power has to be off for a while - like a minute. The event you are showing here is about 10 seconds.
August 22, 201510 yr Author Community Expert I just got a notification from UPS that power had been restored, but I didn't get any notification that it had failed. I know I have gotten power fail notifications before so wonder if this is something that got broken in recent 6.1rc Looking at syslog, it was a very brief event, and I didn't even see any lights dim or anything. Probably time for me to invest in an UPS with AVR. Aug 22 09:17:34 unSERVER dhcpcd[1429]: eth0: xid 0x95612c0a is for hwaddr 00:26:9e:68:b1:46:00:00:00:00:00:00:00:00:00:00 Aug 22 11:37:32 unSERVER apcupsd[11599]: Power failure. Aug 22 11:37:43 unSERVER apcupsd[11599]: Running on UPS batteries. Aug 22 11:37:43 unSERVER apcupsd[11599]: Mains returned. No longer on UPS batteries. Aug 22 11:37:43 unSERVER apcupsd[11599]: Power is back. UPS running on mains. Aug 22 11:37:43 unSERVER sSMTP[29647]: Creating SSL connection to host Aug 22 11:37:43 unSERVER sSMTP[29647]: SSL connection using ECDHE-RSA-AES128-GCM-SHA256 Aug 22 11:37:46 unSERVER sSMTP[29647]: Sent mail for... I haven't tested this is a while, but I don't believe a power failure event is reported that fast. If I remember correctly, the power has to be off for a while - like a minute. The event you are showing here is about 10 seconds. See my post just before yours for a lot more. I think these are actually low voltage events, but they are being put in syslog as Power failure without triggering a notification. When I actually did a test by unplugging the UPS, I got an email in about 3 seconds.
August 22, 201510 yr It's possible it is a brown out. The UPS status on the status page will show if the voltage is low and indicate it is boosting it. I don't think the log and notifications show brown out conditions. I have tested the UPS shutdown on 6.0.1 and 6.1 and the notifications are happening as they are supposed to. I tested and actually found an issue with shutting down the server in 6.1rc5 because of a script change.
August 22, 201510 yr Author Community Expert It's possible it is a brown out. The UPS status on the status page will show if the voltage is low and indicate it is boosting it. I don't think the log and notifications show brown out conditions. I have tested the UPS shutdown on 6.0.1 and 6.1 and the notifications are happening as they are supposed to. I tested and actually found an issue with shutting down the server in 6.1rc5 because of a script change. I think it is working correctly too, and I have had the shutdown work just as it should recently on an actual Power failure. I just was surprised to see these "Power failure" in the syslog and no notification. So I think brownouts are being logged as "Power failure".
Archived
This topic is now archived and is closed to further replies.