On Wed, 14 Apr 2021, J Leslie Turriff wrote:

On 2021-04-14 19:14:01 J Leslie Turriff wrote: This might also be useful: http://rogerprice.org/NUT.html

I havn't done any maintenance on that text for a while. I wrote a more complete NUT User's Guide which is at http://rogerprice.org/NUT/ConfigExamples.A5.pdf On Thu, 15 Apr 2021, Carlos E. R. wrote:

I read the documentation file "/usr/share/doc/packages/nut/README.SUSE"

I try "systemctl start nut-driver.service"

In general one starts NUT by starting the server daemon, and then the monitor daemon. The driver daemon is started automatically by the server. If you have specific questions about the driver, try the NUT mailing list: https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/nut-upsuser Roger

On 15/04/2021 08.41, Roger Price wrote:

On Wed, 14 Apr 2021, J Leslie Turriff wrote:

...

On 2021-04-14 19:14:01 J Leslie Turriff wrote:     This might also be useful: http://rogerprice.org/NUT.html

I havn't done any maintenance on that text for a while.  I wrote a more complete NUT User's Guide which is at http://rogerprice.org/NUT/ConfigExamples.A5.pdf

I'm reading it. Re "Figure 16: NUT provided script for delayed UPS shutdown." ... «This script is executed late in the system shutdown process, and there is no trace in the systemlog of it’s action. If, like the editor, you believe that shutting off power to a system is a majorevent, and should be logged, then you are invited to replace the script provided by NUT with asystemd service unit as shown in appendix 21 which will log the delayed shutdown command.» Why not use "logger"? NAME logger - enter messages into the system log SYNOPSIS logger [options] [message] DESCRIPTION logger makes entries in the system log. When the optional message argument is present, it is written to the log. If it is not present, and the -f option is not given either, then standard input is logged. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 El 2021-04-16 a las 21:48 +0200, Roger Price escribió:

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

...

http://rogerprice.org/NUT/ConfigExamples.A5.pdf

I'm reading it.

Re "Figure 16: NUT provided script for delayed UPS shutdown." ... «This script is executed late in the system shutdown process, and there is no trace in the systemlog of it’s action. If, like the editor, you believe that shutting off power to a system is a majorevent, and should be logged, then you are invited to replace the script provided by NUT with asystemd service unit as shown in appendix 21 which will log the delayed shutdown command.»

Why not use "logger"?

The service unit nut-delayed-ups-shutdown.service described in appendix 21 does use logger:

Ah, didn't look that far yet. But I mean, the script can also use logger, I do all the time. Needs less complication with enabling services and such ;-)

... [Service] Type=oneshot ExecStart=/usr/bin/logger -t nut-delayed-ups-shutdown "upsdrvctl shutting down UPS" ExecStart=/sbin/upsdrvctl shutdown # Debian ...

Sorry about the Debianism - an opensuse sysadmin would replace /sbin/upsdrvctl by /usr/sbin/upsdrvctl

Oh, it is interesting to see the differences :-) - -- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYHnunRwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVgc4AmQF2dYRRdhQlQkgvUj14 XLWCXoEbAJ0UbE6khI2jntMrsEIcLuwzhEx5cw== =9mYu -----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 El 2021-04-16 a las 22:42 +0200, Roger Price escribió:

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

...

...

Why not use "logger"?

The service unit nut-delayed-ups-shutdown.service described in appendix 21 does use logger:

Ah, didn't look that far yet. But I mean, the script can also use logger, I do all the time. Needs less complication with enabling services and such ;-)

The script is executed so late in the shutdown that there is no longer a logging service. That is the reason I suggest the nut-delayed-ups-shutdown service unit.

Ah! - -- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYHn9oRwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfV4P8AniwByxry23XGdwGFi4VY t8Ny7DACAJ9DbniKebS4NOYqv0dqzUnUMyEKnA== =Ztne -----END PGP SIGNATURE-----

Carlos E. R. wrote:

But what command is supposed to start it? The documentation says commands that do not exist!

Enable both nut-driver and nut-monitor services (or start them manually)?

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, 2021-04-15 at 22:25 -0500, David C. Rankin wrote:

On 4/15/21 4:31 AM, Carlos E. R. wrote:

...

But what command is supposed to start it? The documentation says commands that do not exist!

Oh, the nut-server.service E.g.

# systemctl start nut-server

Thanks. Isengard:~ # systemctl start nut-server Isengard:~ # Isengard:~ # systemctl status nut-server ● nut-server.service - Network UPS Tools - power devices information server Loaded: loaded (/usr/lib/systemd/system/nut-server.service; disabled; vendor preset: disabled) Active: active (running) since Fri 2021-04-16 12:08:22 CEST; 1min 2s ago Process: 1244 ExecStart=/usr/sbin/upsd (code=exited, status=0/SUCCESS) Main PID: 1245 (upsd) Tasks: 1 CGroup: /system.slice/nut-server.service └─1245 /usr/sbin/upsd Apr 16 12:08:22 Isengard systemd[1]: Starting Network UPS Tools - power devices information server... Apr 16 12:08:22 Isengard upsd[1244]: fopen /var/lib/ups/upsd.pid: No such file or directory Apr 16 12:08:22 Isengard upsd[1244]: listening on 127.0.0.1 port 3493 Apr 16 12:08:22 Isengard upsd[1244]: listening on 127.0.0.1 port 3493 Apr 16 12:08:22 Isengard upsd[1244]: Connected to UPS [salicru]: blazer_usb-salicru Apr 16 12:08:22 Isengard upsd[1244]: Connected to UPS [salicru]: blazer_usb-salicru Apr 16 12:08:22 Isengard systemd[1]: Started Network UPS Tools - power devices information server. Apr 16 12:08:22 Isengard upsd[1245]: Startup successful Isengard:~ # Well, it does start. nut-driver is started automatically, but not nut-monitor. 3.6> 2021-04-16T12:08:21.783974+02:00 Isengard systemd 1 - - Starting Network UPS Tools - power device driver controller... <3.6> 2021-04-16T12:08:22.063019+02:00 Isengard upsdrvctl 1236 - - Supported UPS detected with megatec protocol <3.6> 2021-04-16T12:08:22.382613+02:00 Isengard upsdrvctl 1236 - - Vendor information unavailable <3.6> 2021-04-16T12:08:22.383145+02:00 Isengard upsdrvctl 1236 - - No values provided for battery high/low voltages in ups.conf <3.6> 2021-04-16T12:08:22.383624+02:00 Isengard upsdrvctl 1236 - - Using 'guestimation' (low: 10.400000, high: 13.000000)! <3.6> 2021-04-16T12:08:22.383976+02:00 Isengard upsdrvctl 1236 - - Battery runtime will not be calculated (runtimecal not set) <3.6> 2021-04-16T12:08:22.640210+02:00 Isengard blazer_usb 1241 - - Startup successful <3.6> 2021-04-16T12:08:22.640909+02:00 Isengard upsdrvctl 1236 - - Network UPS Tools - UPS driver controller 2.7.4 <3.6> 2021-04-16T12:08:22.641365+02:00 Isengard systemd 1 - - Started Network UPS Tools - power device driver controller. <3.6> 2021-04-16T12:08:22.644498+02:00 Isengard systemd 1 - - Starting Network UPS Tools - power devices information server... <3.6> 2021-04-16T12:08:22.652537+02:00 Isengard upsd 1244 - - fopen /var/lib/ups/upsd.pid: No such file or directory <3.6> 2021-04-16T12:08:22.653648+02:00 Isengard upsd 1244 - - listening on 127.0.0.1 port 3493 <3.6> 2021-04-16T12:08:22.654180+02:00 Isengard upsd 1244 - - listening on 127.0.0.1 port 3493 <3.6> 2021-04-16T12:08:22.654630+02:00 Isengard upsd 1244 - - Connected to UPS [salicru]: blazer_usb-salicru <3.6> 2021-04-16T12:08:22.655119+02:00 Isengard upsd 1244 - - Connected to UPS [salicru]: blazer_usb-salicru <3.6> 2021-04-16T12:08:22.658216+02:00 Isengard systemd 1 - - Started Network UPS Tools - power devices information server. <3.6> 2021-04-16T12:08:22.660530+02:00 Isengard upsd 1245 - - Startup successful But I removed the mains power, the UPS switched on, but no message anywhere, no action. I will continue reading. - -- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYHlkhRwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVllYAn3UPVZeoFOIsUgS0MZnj jQGXQRLgAKCK6k1QHtjbr06tfJ3SoqzhoUBuqg== =16+u -----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday, 2021-04-16 at 11:42 +0100, Dave Howorth wrote:

On Fri, 16 Apr 2021 12:18:45 +0200 (CEST) "Carlos E. R." <> wrote:

...

...

Well, it does start.

nut-driver is started automatically, but not nut-monitor.

Didn't somebody say earlier that nut-monitor also had to be started, just like nut-driver? Presumably once enabled and started they'll restart after boot.

I don't know. I also started it manuall, then cycled the mains power. No reaction from nut. [...] Ok, got a reaction. I have to wait several seconds, so probably using "poll". <3.6> 2021-04-16T13:06:19.910036+02:00 Isengard systemd 1 - - Starting Network UPS Tools - power device monitor and shutdown controller... <3.6> 2021-04-16T13:06:19.917370+02:00 Isengard upsmon 2819 - - fopen /var/run/upsmon.pid: No such file or directory <3.6> 2021-04-16T13:06:19.918869+02:00 Isengard upsmon 2819 - - UPS: salicru@127.0.0.1:3493 (master) (power value 1) <3.6> 2021-04-16T13:06:19.919287+02:00 Isengard systemd 1 - - nut-monitor.service: PID file /var/run/upsmon.pid not readable (yet?) after start: No such file or directory <3.6> 2021-04-16T13:06:19.919900+02:00 Isengard upsmon 2820 - - Startup successful <3.6> 2021-04-16T13:06:19.920448+02:00 Isengard upsmon 2819 - - Using power down flag file /etc/killpower <3.4> 2021-04-16T13:06:19.920683+02:00 Isengard systemd 1 - - nut-monitor.service: Supervising process 2821 which is not our child. We'll most likely not notice when it exits. <3.6> 2021-04-16T13:06:19.921333+02:00 Isengard systemd 1 - - Started Network UPS Tools - power device monitor and shutdown controller. <3.6> 2021-04-16T13:06:19.922519+02:00 Isengard upsd 1245 - - User upsmon@127.0.0.1 logged into UPS [salicru] <3.5> 2021-04-16T13:06:59.926598+02:00 Isengard upsmon 2821 - - UPS salicru@127.0.0.1:3493 on battery <3.5> 2021-04-16T13:07:19.928924+02:00 Isengard upsmon 2821 - - UPS salicru@127.0.0.1:3493 on line power Well, it is "working" now. No battery status information? (I need more reading) - -- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYHlyKhwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVEtUAn2EPKNerfZNsmWGU9uI3 tNsuDgT6AJ9tcN8MzyVwkhr4FZ2BZaqkGwz3Eg== =hkRA -----END PGP SIGNATURE-----

On 16/04/2021 14.08, Roger Price wrote:

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

...

Didn't somebody say earlier that nut-monitor also had to be started, just like nut-driver? Presumably once enabled and started they'll restart after boot.

Yes, they both need starting.  The server first.

...

No battery status information? (I need more reading)

Independent of the monitor, command upsc <myUPS> will give the current status of the UPS.

It does now, but until today it did not. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 El 2021-04-16 a las 20:25 +0100, Dave Howorth escribió:

On Fri, 16 Apr 2021 14:08:17 +0200 (CEST) Roger Price wrote:

...

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

...

Didn't somebody say earlier that nut-monitor also had to be started, just like nut-driver? Presumably once enabled and started they'll restart after boot.

Yes, they both need starting. The server first.

So there should be an After or a Before dependency somewhere in the service files?

Correct, the nut-server.service file "wants" nut-driver.service. cer@Telcontar:~> systemctl cat nut-server.service # /usr/lib/systemd/system/nut-server.service [Unit] Description=Network UPS Tools - power devices information server After=local-fs.target network.target nut-driver.service # We don't Require drivers to be successfully started! This would be # a change of behavior compared to init SysV, and could prevent from # accessing successfully started, at least to audit a system. Wants=nut-driver.service <========== Before=nut-monitor.service <---------- [Service] ExecStart=/usr/sbin/upsd Type=forking [Install] WantedBy=multi-user.target cer@Telcontar:~> cer@Telcontar:~> systemctl cat nut-monitor.service # /usr/lib/systemd/system/nut-monitor.service [Unit] Description=Network UPS Tools - power device monitor and shutdown controller After=local-fs.target network.target nut-server.service [Service] ExecStart=/usr/sbin/upsmon PIDFile=/var/run/upsmon.pid Type=forking [Install] WantedBy=multi-user.target cer@Telcontar:~> cer@Telcontar:~> systemctl cat nut-driver.service # /usr/lib/systemd/system/nut-driver.service [Unit] Description=Network UPS Tools - power device driver controller After=local-fs.target network.target StopWhenUnneeded=yes [Service] ExecStart=/usr/sbin/upsdrvctl start ExecStop=/usr/sbin/upsdrvctl stop Type=forking cer@Telcontar:~> - -- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYHnnHhwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVL+0Ani2KQrlTIr7BfgSYpXyh pd7EmJVcAJ9MruAAfu1xyU3gioHt/xEH5GU25A== =D8QZ -----END PGP SIGNATURE-----

On Fri, 16 Apr 2021, David C. Rankin wrote:

On 4/16/21 5:42 AM, Dave Howorth wrote:

...

...

Well, it does start.

nut-driver is started automatically, but not nut-monitor. Didn't somebody say earlier that nut-monitor also had to be started, just like nut-driver? Presumably once enabled and started they'll restart after boot.

[snip]

You do NOT need nut-monitor. That is for monitoring additional peripherals attached to a nut server, e.g.

I assume that by nut-monitor you mean program upsmon. This is the daemon which takes the decision to shutdown a system when it reaches status OB-LB (on battery and low battery). If you do not have this program, how will you order the shutdown?

NUT-Monitor is a graphical user interface to monitor and manage devices connected to the Network UPS Tools server.

Are you referring to upsmon or some other program? Roger

On 4/16/21 2:01 PM, Carlos E. R. wrote:

Without nut-monitor, which is a service, not a GUI, there are no messages in the log when I disconnect the mains. The entire thing becomes useless, there is no reaction to a power failure.

On the other hand, nut is not correctly integrated with systemd:

Apologies, I was thinking about the web interface cgi scripts. Yes, you need to part that shuts the server down. -- David C. Rankin, J.D.,P.E.

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday, 2021-04-16 at 12:45 -0500, David C. Rankin wrote:

On 4/16/21 5:18 AM, Carlos E. R. wrote:

...

On Thursday, 2021-04-15 at 22:25 -0500, David C. Rankin wrote:

...

On 4/15/21 4:31 AM, Carlos E. R. wrote:

...

But what command is supposed to start it? The documentation says commands that do not exist!

Oh, the nut-server.service  E.g.

# systemctl start nut-server

Thanks.

Also "nut-monitor.service" has to be started and enabled.

Now just,

$ upsc blazer_usb-salicru

to see the UPS details.

Yes, I noticed, after sucessfully starting the above services. Isengard:~ # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 device.type: ups driver.name: blazer_usb driver.parameter.pollinterval: 2 driver.parameter.port: auto driver.parameter.synchronous: no driver.version: 2.7.4 driver.version.internal: 0.12 input.current.nominal: 3.0 input.frequency: 50.0 input.frequency.nominal: 50 input.voltage: 226.9 input.voltage.fault: 226.9 input.voltage.nominal: 230 output.voltage: 226.9 ups.beeper.status: enabled ups.delay.shutdown: 30 ups.delay.start: 180 ups.load: 10 ups.productid: 5161 ups.status: OL ups.temperature: 25.0 ups.type: offline / line interactive ups.vendorid: 0665 Isengard:~ # But I don't see anything yet that collects stats. - -- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYHngURwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVkXwAn2Ak6K4+PMonbHK2vBDv KX1+3eK8AJ0cqc4Zh8a9ipWLGRmIeN0He0YuVQ== =uczE -----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 El 2021-04-16 a las 21:29 +0200, Roger Price escribió:

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

Isengard:~ # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 ...

I don't see variable battery.charge.low which specifies the charge below which the LB (Low Battery) status is raised, signalling that the time has come to shutdown.

Does the command "upsrw Salicru" list the variable battery.charge.low ?

Isengard:~ # upsrw Salicru Isengard:~ # Nope... - -- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYHnnshwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVmncAn3Poncj1osM54g3/QDa6 fk41qye8AKCIdB0Yd3LWGJ0aAzE+1ZiumA82ow== =pTVt -----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Friday, 2021-04-16 at 22:22 +0200, Roger Price wrote:

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

El 2021-04-16 a las 21:29 +0200, Roger Price escribió:

...

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

Isengard:~ # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 ...

I don't see variable battery.charge.low which specifies the charge below which the LB (Low Battery) status is raised, signalling that the time has come to shutdown.

Does the command "upsrw Salicru" list the variable battery.charge.low ?

Isengard:~ # upsrw Salicru Isengard:~ #

Nope...

If you are using the blazer_usb driver there is a workaround using extra arguments to the driver:

* default.battery.voltage.high = value

Maximum battery voltage that is reached after about 12 to 24 hours charging. If you want the driver to report a guesstimated battery.charge, you need to specify this (see BATTERY CHARGE).

Well, that voltage is the same for every battery of the same chemicals :-)

* default.battery.voltage.low = value

Minimum battery voltage just before the UPS automatically shuts down. If you want the driver to report a guesstimated battery.charge, you need to specify this (see BATTERY CHARGE).

I don't like the idea of stressing the battery to exhaustion in order to find out that value. Doesn't the hardware communicate the battery percent itself? It has an LCD display with a graph display of charge. I have this: Isengard:~ # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 Aren't those the values told by the UPS itself? It prints battery charge.

See man blazer_usb

"Salicru" is not even listed at https://networkupstools.org/stable-hcl.html There is only a user report for the "SPS One 700VA": https://networkupstools.org/ddl/Salicru/index.html Anyway... I changed the configuration a bit: [salicru] driver = blazer_usb #port = /dev/usb/hiddev0 port = auto desc = "Salicru SPS SOHO+ 800VA" default.battery.voltage.high = 13.60 default.battery.voltage.low = 10.40 The utility reports the same values as before: Isengard:/etc/ups # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 device.type: ups driver.name: blazer_usb I switched off the mains for a few seconds: sengard:/etc/ups # upsc Salicru battery.charge: 92 <============= battery.voltage: 12.80 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 device.type: ups driver.name: blazer_usb driver.parameter.pollinterval: 2 driver.parameter.port: auto driver.parameter.synchronous: no driver.version: 2.7.4 driver.version.internal: 0.12 input.current.nominal: 3.0 input.frequency: 50.0 input.frequency.nominal: 50 input.voltage: 3.9 <=========== input.voltage.fault: 3.9 input.voltage.nominal: 230 output.voltage: 227.7 ups.beeper.status: enabled ups.delay.shutdown: 30 ups.delay.start: 180 ups.load: 6 ups.productid: 5161 ups.status: OB ups.temperature: 25.0 ups.type: offline / line interactive ups.vendorid: 0665 Isengard:/etc/ups # And with mains back on: Isengard:/etc/ups # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 device.type: ups driver.name: blazer_usb driver.parameter.pollinterval: 2 driver.parameter.port: auto driver.parameter.synchronous: no driver.version: 2.7.4 driver.version.internal: 0.12 input.current.nominal: 3.0 input.frequency: 50.0 input.frequency.nominal: 50 input.voltage: 228.9 input.voltage.fault: 228.9 input.voltage.nominal: 230 output.voltage: 228.9 ups.beeper.status: enabled ups.delay.shutdown: 30 ups.delay.start: 180 ups.load: 7 ups.productid: 5161 ups.status: OL ups.temperature: 25.0 ups.type: offline / line interactive ups.vendorid: 0665 Isengard:/etc/ups # - -- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar) -----BEGIN PGP SIGNATURE----- iHoEARECADoWIQQZEb51mJKK1KpcU/W1MxgcbY1H1QUCYHn3xBwccm9iaW4ubGlz dGFzQHRlbGVmb25pY2EubmV0AAoJELUzGBxtjUfVqR0AnAyIGssvXhAt6NsrtiRi LghqlteTAJ4qT4htx32MSZZhda7R2II4OY6cqw== =UIlv -----END PGP SIGNATURE-----

On 4/16/21 3:47 PM, Carlos E. R. wrote:

On Friday, 2021-04-16 at 22:22 +0200, Roger Price wrote:

...

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

El 2021-04-16 a las 21:29 +0200, Roger Price escribió:

...

On Fri, 16 Apr 2021, Carlos E. R. wrote:

...

Isengard:~ # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 ...

I don't see variable battery.charge.low which specifies the charge below which the LB (Low Battery) status is raised, signalling that the time has come to shutdown.

Does the command "upsrw Salicru" list the variable battery.charge.low ?

Isengard:~ # upsrw Salicru Isengard:~ # Nope...

...

If you are using the blazer_usb driver there is a workaround using extra arguments to the driver:

...

* default.battery.voltage.high = value

...

Maximum battery voltage that is reached after about 12 to 24 hours charging. If you want the driver to report a guesstimated battery.charge, you need to specify this (see BATTERY CHARGE).

Well, that voltage is the same for every battery of the same chemicals :-)

...

* default.battery.voltage.low = value

...

Minimum battery voltage just before the UPS automatically shuts down. If you want the driver to report a guesstimated battery.charge, you need to specify this (see BATTERY CHARGE).

I don't like the idea of stressing the battery to exhaustion in order to find out that value. Doesn't the hardware communicate the battery percent itself? It has an LCD display with a graph display of charge.

I have this:

Isengard:~ # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0

Aren't those the values told by the UPS itself? It prints battery charge.

...

See man blazer_usb

"Salicru" is not even listed at https://networkupstools.org/stable-hcl.html

There is only a user report for the "SPS One 700VA":

https://networkupstools.org/ddl/Salicru/index.html

Anyway...

I changed the configuration a bit:

[salicru] driver = blazer_usb #port = /dev/usb/hiddev0 port = auto desc = "Salicru SPS SOHO+ 800VA" default.battery.voltage.high = 13.60 default.battery.voltage.low = 10.40

The utility reports the same values as before:

Isengard:/etc/ups # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 device.type: ups driver.name: blazer_usb

I switched off the mains for a few seconds:

sengard:/etc/ups # upsc Salicru battery.charge: 92 <============= battery.voltage: 12.80 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 device.type: ups driver.name: blazer_usb driver.parameter.pollinterval: 2 driver.parameter.port: auto driver.parameter.synchronous: no driver.version: 2.7.4 driver.version.internal: 0.12 input.current.nominal: 3.0 input.frequency: 50.0 input.frequency.nominal: 50 input.voltage: 3.9 <=========== input.voltage.fault: 3.9 input.voltage.nominal: 230 output.voltage: 227.7 ups.beeper.status: enabled ups.delay.shutdown: 30 ups.delay.start: 180 ups.load: 6 ups.productid: 5161 ups.status: OB ups.temperature: 25.0 ups.type: offline / line interactive ups.vendorid: 0665 Isengard:/etc/ups #

And with mains back on:

Isengard:/etc/ups # upsc Salicru battery.charge: 100 battery.voltage: 13.60 battery.voltage.high: 13.00 battery.voltage.low: 10.40 battery.voltage.nominal: 12.0 device.type: ups driver.name: blazer_usb driver.parameter.pollinterval: 2 driver.parameter.port: auto driver.parameter.synchronous: no driver.version: 2.7.4 driver.version.internal: 0.12 input.current.nominal: 3.0 input.frequency: 50.0 input.frequency.nominal: 50 input.voltage: 228.9 input.voltage.fault: 228.9 input.voltage.nominal: 230 output.voltage: 228.9 ups.beeper.status: enabled ups.delay.shutdown: 30 ups.delay.start: 180 ups.load: 7 ups.productid: 5161 ups.status: OL ups.temperature: 25.0 ups.type: offline / line interactive ups.vendorid: 0665 Isengard:/etc/ups #

-- Cheers, Carlos E. R. (from openSUSE 15.2 x86_64 at Telcontar)

The battery.charge.low is just the percentage of battery you want remaining when the shutdown is initiated. Usually it defaults to 10 (10% remaining). You do want to ensure you have it set, either by hook, crook or hack, e.g. # upsc phoinix_ups battery.charge: 100 battery.charge.low: 10 battery.charge.warning: 20 battery.mfr.date: CPS battery.runtime: 1710 battery.runtime.low: 300 battery.type: PbAcid battery.voltage: 24.0 battery.voltage.nominal: 24 device.mfr: CPS device.model: CP1350PFCLCD device.serial: 000000000000 device.type: ups driver.name: usbhid-ups driver.parameter.pollfreq: 30 driver.parameter.pollinterval: 2 driver.parameter.port: /dev/hiddev0 driver.parameter.synchronous: no driver.version: 2.7.4 driver.version.data: CyberPower HID 0.4 driver.version.internal: 0.41 input.transfer.high: 139 input.transfer.low: 88 input.voltage: 125.0 input.voltage.nominal: 120 output.voltage: 141.0 ups.beeper.status: enabled ups.delay.shutdown: 20 ups.delay.start: 30 ups.load: 22 ups.mfr: CPS ups.model: CP1350PFCLCD ups.productid: 0501 ups.realpower.nominal: 810 ups.serial: 000000000000 ups.status: OL ups.test.result: No test initiated ups.timer.shutdown: -60 ups.timer.start: -60 ups.vendorid: 0764 Nut never got around to dividing the battery.voltage by 2 for this 2-12V battery UPS. -- David C. Rankin, J.D.,P.E.

On Sat, 17 Apr 2021, Carlos E. R. wrote:

Methinks that depleting the battery to 10 is a very severe wear on it. I would go for 50% tops. If the system is not capable of going down fast...

If you try to power down my machine at 23:50 hours, it will not go down till 00:15, or worse, 00:30. Other times, yes, it will go down fast.

How do you hold back /sbin/shutdown for what might be 40 minutes? Is the Salicru capable of supplying power to the system for 40 minutes and still have 50% reserve? This will need testing. Roger

On Sat, 17 Apr 2021, Carlos E. R. wrote:

On 17/04/2021 21.43, Roger Price wrote:

...

On Sat, 17 Apr 2021, Carlos E. R. wrote:

...

Methinks that depleting the battery to 10 is a very severe wear on it. I would go for 50% tops. If the system is not capable of going down fast...

If you try to power down my machine at 23:50 hours, it will not go down till 00:15, or worse, 00:30. Other times, yes, it will go down fast.

How do you hold back /sbin/shutdown for what might be 40 minutes?  Is the Salicru capable of supplying power to the system for 40 minutes and still have 50% reserve?  This will need testing.

I don't. If the mains power fails that instant and I'm not there, the UPS dies of exhaustion and the machine with it.

For me, this would not be acceptable.

At least with hibernation, which is what I do. The system tries to sync the filesystem, and if it does that during the interval 23:45 to 00:15 the kernel is not capable of syncing the filesystem while texpire is running. The cron-job texpire has to be killed first, then the machine can be told to hibernate. I have not tested this yet.

What is the advantage of hibernation for a desktop system? The machine does no useful work, yet consumes energy from the UPS. A "proper" shutdown would kill texpire and not drain the battery, which you have said you want to avoid. I do not use hibernation. I find I am better served with a shutdown --poweroff and an automatic restart when utility power returns.

I have not tested power down during that interval yet (I thought I had).

There are other issues:

- sometimes hibernation fails, and has to be called a second time. How to automate this?

I'm guessing that you specify something like SHUTDOWNCMD "systemctl start systemd-hibernate.service" in upsmon.conf. You would have to replace this with a call of a script which watches over systemctl status systemd-hibernate.service and restarts the service if it fails. This is becoming complicated. It might also be advisable to not execute the command upsdrvctl shutdown in order to avoid turning off the UPS power outlets. Roger

On 4/17/21 6:24 AM, Carlos E. R. wrote:

Methinks that depleting the battery to 10 is a very severe wear on it. I would go for 50% tops. If the system is not capable of going down fast...

If you try to power down my machine at 23:50 hours, it will not go down till 00:15, or worse, 00:30. Other times, yes, it will go down fast.

Eh..., Normally I agree, but here you are just handling the power fail case. I always get to my machines and shut them down before they reach 10%. Only time they exhaust the battery to 10% is if the office goes down at night -- and well, that's what the $25 batteries are supposed to do. Regardless of the number of discharges, I always get 3-5 years out of the batteries. I have a separate UPS that only powers the cable modem and wireless router, I will let go to the 10% for any outage that lasts that long. Generally gives about 2.5 hours of runtime. Setting at 50% would be fine, that would just shorten the runtime to power off. Whatever you set it at is fine -- just make sure it is set :) -- David C. Rankin, J.D.,P.E.

On 4/17/21 4:35 PM, Carlos E. R. wrote:

The other day the mains power failed (the house earth fault switch triggered), and the mini server went down, its UPS could not take the load for maybe the 2 minutes it took me to notice and reset the switch. The other two UPS in the house took the load just fine.

Chuckling... You recall the ice storm in Texas if February? I was without power for 6 days 6 hours and the temperature got down to -3 F (41 inside...). No UPS on earth would have saved me :) -- David C. Rankin, J.D.,P.E.