Smart UPS 420 fails with system overload repeatable on saturday afternoon between 3 and 4 a clock pm

[ecaroh_apc]

I have an SMART UPS 420 with batteries one year old. There should be no battery overload, cause besides a modem and a router only one small server is attached. APCUPSD says it has 33 % UPS Load, 100 % Battery capacity and 32 mins run time remaining.

The UPS suddenly fails supplying power. The server and the modem are powered of. The UPS has an ongoing beep and the system overload LED is on. I have to switch the UPS on again.

The surprising fact is, that this happened round about 5 times in the last year always on a saturday afternoon between 3 and 4 pm. There are no hints from the apcupsd event logs. I have no clue where to start, cause i can not imagine an overload causing the situation. Especially when there should be not more load on the weekend. The monitoring of the server has no hint of extra cpu load at this time. And there has been more load on it while it worked without outage. To reduce it i bought a second UPS and divided system load between them. 

Below the output of apcupsd running on linux with serial cable. Does anybody have ideas of a saturday afternoon overload? The last outage happend apr. at 4:00 pm 1 and a half hour before the last log line on 2015-11-21 17:23:38 with message unknown startup succeeded when i bring back the ups online again.

Just a guess; can apcupsd on linux shut down the power via the serial cable? Then it maybe a fault on apcupsd side?

This are samples of "unknown startup succeeded" - always saturday after a shutdown/overload where i started the ups again:

  1 2015-11-21 17:23:38 +0100

  2 2015-10-24 16:56:27 +0200

  3 2015-08-29 17:39:41 +0200

  4 2015-07-18 09:20:09 +0200

  5 2015-07-04 11:22:44 +0200

  6 2015-03-21 13:18:41 +0100

  7 2015-01-17 15:28:47 +0100

  8 2014-12-06 15:20:39 +0100

  9 2014-11-15 16:47:59 +0100

10 2014-11-08 17:28:59 +0100

11 2014-10-11 17:20:36 +0200

Monitoring: of-srv-ipf-1
 UPS Model: Smart-UPS 420   
  UPS Name: UPS_IDEN
   APCUPSD: Version 3.14.10
    Status: ONLINE 

Last UPS Self Test: NO
Last Test Date: Not found

Utility Voltage: 233.2 VAC
   Line Minimum: 231.8 VAC
   Line Maximum: 233.2 VAC
    Output Freq: 50.0 Hz

2015-11-21 17:23:38 +0100 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-10-24 16:56:27 +0200 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-09-17 15:58:20 +0200 Power is back. UPS running on mains.
2015-09-17 15:58:18 +0200 Power failure.
2015-08-29 17:39:41 +0200 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-08-29 16:31:06 +0200 apcupsd shutdown succeeded
2015-08-29 16:31:06 +0200 apcupsd exiting, signal 15
2015-08-29 14:52:36 +0200 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-08-29 14:49:06 +0200 apcupsd shutdown succeeded
2015-08-29 14:49:06 +0200 apcupsd exiting, signal 15
2015-08-25 15:39:32 +0200 Power is back. UPS running on mains.
2015-08-25 15:39:30 +0200 Power failure.
2015-08-15 16:22:47 +0200 UPS Self Test completed: Battery OK
2015-08-15 16:22:39 +0200 UPS Self Test switch to battery.
2015-08-01 16:22:13 +0200 UPS Self Test completed: Battery OK
2015-08-01 16:22:06 +0200 UPS Self Test switch to battery.
2015-07-24 14:44:00 +0200 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-07-24 14:39:54 +0200 apcupsd shutdown succeeded
2015-07-24 14:39:54 +0200 apcupsd exiting, signal 15
2015-07-18 16:22:30 +0200 UPS Self Test completed: Battery OK
2015-07-18 16:22:27 +0200 UPS Self Test switch to battery.
2015-07-18 09:20:09 +0200 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-07-18 09:16:43 +0200 apcupsd shutdown succeeded
2015-07-18 09:16:43 +0200 apcupsd exiting, signal 15
2015-07-04 11:22:44 +0200 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-07-04 11:19:17 +0200 apcupsd shutdown succeeded
2015-07-04 11:19:17 +0200 apcupsd exiting, signal 15
2015-06-01 03:02:31 +0200 Power is back. UPS running on mains.
2015-06-01 03:02:29 +0200 Power failure.
2015-06-01 03:01:47 +0200 Power is back. UPS running on mains.
2015-06-01 03:01:45 +0200 Power failure.
2015-04-01 17:24:35 +0200 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-04-01 17:21:03 +0200 apcupsd shutdown succeeded
2015-04-01 17:21:03 +0200 apcupsd exiting, signal 15
2015-03-21 13:18:41 +0100 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-03-21 13:15:17 +0100 apcupsd shutdown succeeded
2015-03-21 13:15:17 +0100 apcupsd exiting, signal 15
2015-01-28 11:11:27 +0100 Power is back. UPS running on mains.
2015-01-28 11:11:26 +0100 Power failure.
2015-01-25 12:40:48 +0100 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-01-25 12:37:30 +0100 apcupsd shutdown succeeded
2015-01-25 12:37:30 +0100 apcupsd exiting, signal 15
2015-01-17 15:28:47 +0100 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-01-14 19:23:09 +0100 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2015-01-14 19:19:46 +0100 apcupsd shutdown succeeded
2015-01-14 19:19:46 +0100 apcupsd exiting, signal 15
2014-12-20 15:17:33 +0100 UPS Self Test completed: Battery OK
2014-12-20 15:17:26 +0100 UPS Self Test switch to battery.
2014-12-06 15:20:39 +0100 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded
2014-11-15 16:47:59 +0100 apcupsd 3.14.10 (13 September 2011) unknown startup succeeded


output of apcupsd.conf without comments: 

UPSCABLE smart

UPSTYPE apcsmart

DEVICE /dev/ttyS0

LOCKFILE /var/lock

SCRIPTDIR /etc/apcupsd

PWRFAILDIR /etc/apcupsd

NOLOGINDIR /etc

ONBATTERYDELAY 6

BATTERYLEVEL 5

MINUTES 3

TIMEOUT 0

ANNOY 300

ANNOYDELAY 60

NOLOGON disable

KILLDELAY 0

NETSERVER on

NISIP 0.0.0.0

NISPORT 3551

EVENTSFILE /var/log/apcupsd.events

EVENTSFILEMAX 10

UPSCLASS standalone

UPSMODE disable

STATTIME 0

STATFILE /var/log/apcupsd.status

LOGSTATS off

DATATIME 0