[Nut-upsuser] System with MGE UPS shuts down too early

[Georgi D. Sotirov]

Hello,

I'm experiencing a strange early shutdown behavior with the following 
system:

  * */OS/*: Slackware Linux 14.2 x86_64
  * */NUT/*: 2.7.4 (from my own package
    <https://sotirov-bg.net/slackpack/pack.cgi?id=4083>)
  * */UPS/*: MGE Ellipse PRO 1600
    <powerquality.eaton.com/Products-services/Backup-Power-UPS/Ellipse-PRO.aspx>
    (bought in September 2017)

*/Problem description/*: An hour ago there was a power outage for about 
3-4 minutes. However, the system was forcefully shut down just after a 
minute (!) and I'm not sure I understand why. This is what I found into 
the logs:

  * /var/log/messages (only relevant):

Jan 11 11:49:15 sotirov-bg upsmon[2677]: UPS mgeups at localhost on battery
Jan 11 11:49:15 sotirov-bg upssched[29746]: Executing command: 
ups-on-battery
Jan 11 11:49:15 sotirov-bg upssched-cmd: Unrecognized command: 
ups-on-battery
Jan 11 11:50:20 sotirov-bg upsmon[2677]: UPS mgeups at localhost battery is low
Jan 11 11:50:20 sotirov-bg upsd[2672]: Client monuser at 127.0.0.1 set FSD 
on UPS [mgeups]
Jan 11 11:50:20 sotirov-bg upsmon[2677]: Auto logout and shutdown proceeding
Jan 11 11:50:20 sotirov-bg upssched[30461]: Executing command: ups-low-batt
Jan 11 11:50:20 sotirov-bg upssched-cmd: Unrecognized command: ups-low-batt
Jan 11 11:50:20 sotirov-bg upssched[30466]: Executing command: ups-shutdown
Jan 11 11:50:20 sotirov-bg upssched-cmd: Unrecognized command: ups-shutdown
Jan 11 11:50:25 sotirov-bg shutdown[30471]: shutting down for system halt
Jan 11 11:50:25 sotirov-bg init: Switching to runlevel: 0

  * /var/log/ups (snippet of consecutive messages):

20200111 114901 100 221.0 17 [OL] NA 50.0
20200111 114906 100 221.0 17 [OL] NA 50.0
20200111 114911 100 221.0 17 [OL] NA 50.0
20200111 114916 99 0.0 19 [OB] NA 0.0
20200111 114921 99 0.0 19 [OB] NA 0.0
20200111 114926 99 0.0 19 [OB] NA 0.0
20200111 114931 99 0.0 19 [OB] NA 0.0
20200111 114936 99 0.0 19 [OB] NA 0.0
20200111 114941 99 0.0 19 [OB] NA 0.0
20200111 114946 98 0.0 16 [OB] NA 0.0
20200111 114951 98 0.0 16 [OB] NA 0.0
20200111 114956 98 0.0 16 [OB] NA 0.0
20200111 115001 98 0.0 16 [OB] NA 0.0
20200111 115006 98 0.0 16 [OB] NA 0.0
20200111 115011 98 0.0 16 [OB] NA 0.0
20200111 115016 98 0.0 16 [OB] NA 0.0
20200111 115021 96 0.0 18 [FSD ALARM OB LB] NA 0.0
20200111 115026 96 0.0 18 [FSD ALARM OB LB] NA 0.0
20200111 115031 96 0.0 18 [FSD ALARM OB LB] NA 0.0
20200111 115036 96 0.0 18 [FSD ALARM OB LB] NA 0.0
20200111 115041 96 0.0 18 [FSD ALARM OB LB] NA 0.0
20200111 115044 96 0.0 18 [FSD ALARM OB LB] NA 0.0
20200111 120209 NA NA NA [WAIT] NA NA
20200111 120214 84 217.0 16 [OL] NA 50.0
20200111 120219 84 217.0 16 [OL] NA 50.0
20200111 120224 84 217.0 16 [OL] NA 50.0

What bugs me is that for just over a minute on battery power the status 
changes to FSD ALARM OB LB (see in red above), while at this moment the 
battery charge is 96 % (!?). After the power was restored the system 
didn't come up, so I had to power it up manually. Checking the battery 
charge afterwards it was still 85 %, so I'm really not sure why NUT 
shutdown the system so early. I have tested the shutdown sequence in 
September 2017 after I installed the UPS by manually removing the online 
power and the system was shutdown properly after about 30 minutes. In 
several previous occasions I didn't notice such problem and the system 
was shutdown after the battery reached the critical charge level of 20% 
or managed to discharge the battery until online power restored.

Any suggestions what could be the problem? What does this "ALARM" in 
status mean? I checked older ups monitor logs and I did not find such 
status.


Best regards,
--
Georgi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/nut-upsuser/attachments/20200111/e2e00409/attachment.html>