On Sat, Mar 29, 2025, at 5:24 PM, Dan Langille via Nut-upsuser wrote: > > > On Sat, Mar 29, 2025, at 4:40 PM, Jim Klimov wrote: >> The problem with computers - they do what you tell them to, not what you >> might want them to. >>
I've never liked them. Terrible things. It's merely a fad. >> Here your heartbeat.dev literally tells the emulation driver to set >> ups.status=OL, wait 300s, set it to OB, wait 300s, repeat. >> >> So NUT reporting that the heartbeat UPS became on-battery 5 minutes after >> start means these pieces are doing their job correctly and talking to each >> other well. >> >> Jim >> > > Outside eyes are so useful. > > I knew what you meant as soon as I read the second paragraph. After going through the original post (https://alioth-lists.debian.net/pipermail/nut-upsuser/2017-April/010590.htm) again, I now understand how it's trying to achieve the goal. I also managed to fix several other issues with my configuration. NOTE to readers: don't use my postedconfiguration it was wrong. For example, most of my NOTIFYCMD should have been NOTIFYFLAG (I suspect copy/paste failures). I also corrected upsd.users and upsmon.conf which had conflicting users. I later discovered that heartbeat.dev must now be heartbeat.conf - the former partially works. Details below. Here is an example of restarting nut while nut_upsmon is running. Seems sensible to me. Mar 30 20:05:58 gw01 upsd[73103]: mainloop: Interrupted system call Mar 30 20:06:00 gw01 upsmon[76592]: Poll UPS [ups04] failed - Server disconnected Mar 30 20:06:00 gw01 upsmon[76592]: Communications with UPS ups04 lost Mar 30 20:06:00 gw01 upsmon[76592]: Poll UPS [heartbeat] failed - Write error: Broken pipe Mar 30 20:06:00 gw01 upsmon[76592]: Communications with UPS heartbeat lost Mar 30 20:06:05 gw01 upsmon[76592]: UPS [ups04]: connect failed: Connection failure: Connection refused Mar 30 20:06:05 gw01 upsmon[76592]: UPS [heartbeat]: connect failed: Connection failure: Connection refused Mar 30 20:06:10 gw01 upsmon[76592]: UPS [ups04]: connect failed: Connection failure: Connection refused Mar 30 20:06:10 gw01 upsmon[76592]: UPS [heartbeat]: connect failed: Connection failure: Connection refused Mar 30 20:06:15 gw01 upsmon[76592]: Communications with UPS ups04 established Mar 30 20:06:15 gw01 upsmon[76592]: Communications with UPS heartbeat established Mar 30 20:11:05 gw01 upsmon[76592]: UPS heartbeat on battery However, I never see a switch back to line power. I've waited about 35 minutes now. Reading further, I rediscovered: https://rogerprice.org/NUT/ConfigExamples.A5.pdf In there, I found: 2020-08-20 File heartbeat.dev becomes heartbeat.conf I renamed all instances of heartbeat.dev to heartbeat.conf, and restarted both nut & upsmon. Boom. Actual results matched expected results: Mar 30 20:54:22 gw01 upsmon[76592]: UPS heartbeat on battery Mar 30 20:59:23 gw01 upsmon[76592]: UPS heartbeat on line power Mar 30 21:04:24 gw01 upsmon[76592]: UPS heartbeat on battery Mar 30 21:09:29 gw01 upsmon[76592]: UPS heartbeat on line power Mar 30 21:14:30 gw01 upsmon[76592]: UPS heartbeat on battery Mar 30 21:19:30 gw01 upsmon[76592]: UPS heartbeat on line power Mar 30 21:24:31 gw01 upsmon[76592]: UPS heartbeat on battery Mar 30 21:29:37 gw01 upsmon[76592]: UPS heartbeat on line power Mar 30 21:34:37 gw01 upsmon[76592]: UPS heartbeat on battery To verify it was the device name, I stopped both daemons and made the following modifications: * ups.conf: 'port = heartbeat.conf' -> 'port = heartbeat.dev' * mv heartbeat.conf heartbeat.conf Start nut, start ups_mon. With that configuration, the previously mentioned behavior returned. Reverting the changes resulted in the expected swapping between battery and line every 5 minutes. Thank you for getting me started on the right track. -- Dan Langille [email protected]
_______________________________________________ Nut-upsuser mailing list [email protected] https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/nut-upsuser
