NUT - fopen & excessive poll failures

[Leander]

Something seems to go wrong with NUT, and I assume it might be the rc.d script of NUT.
Also the "excessive poll failures" seem weired to me, cause I have the default "pollfreq = 30". Can you explain this strange behaviour?

#
# ==== Result of service nut start and service nut_upsmon start ====
#

Detected Smart-UPS RT 2000 XL on host 192.168.50.5 (mib: apcc 1.2)
[RT2000XL] Warning: excessive poll failures, limiting error reporting
[RT2000XL] Warning: excessive poll failures, limiting error reporting
Failed to read pid from /var/db/nut/upsd.pid
listening on 127.0.0.1 port 3493
listening on ::1 port 3493
Connected to UPS [RT2000XL]: snmp-ups-RT2000XL

Network UPS Tools - UPS driver controller 2.7.3
Starting nut.
fopen /var/db/nut/upsmon.pid: No such file or directory
Using power down flag file /etc/killpower
UPS: RT2000XL@localhost (master) (power value 1)

Starting nut_upsmon.

#
# ==== Config ====
#

hosts.conf

MONITOR RT2000XL@localhost "Smart-UPS RT 2000 XL"

ups.conf

[RT2000XL]
  desc = "Smart-UPS RT 2000 XL"
  driver = snmp-ups
  port = 192.168.0.133
  mibs = apcc
  pollfreq = 30
  community = public
  snmp_version = v1

upsd.users

# Supervision user
[admin]
  password = PASSWORD
  actions  = SET
  instcmds = ALL

# Protection user
[monuser]
  password  = PASSWORD
  upsmon master

upsmon.conf

MONITOR RT2000XL@localhost 1 monuser PASSWORD master

• FreeBSD 10.2-RELEASE
  
• NUT Version 2.7.3_2

 

[junovitch@]

It could be specific to that model. Can you try sh -x /usr/local/etc/rc.d/nut start and post that output to a site like pastebin? Or just post the couple lines before and after the "excessive poll failure" in a [CODE][/CODE] block here. Let's try to get a better idea which command the startup script runs that is causing the issue.