[Nut-upsdev] more on 2.0.3-pre

[Peter Selinger]

Hi Alfred,

thanks for your testing. I also have a Back-UPS ES 650, but it is
currently in my wife's office at work so I am not doing any testing
with it.

Thanks for catching the writable variables. That is of course useful
to make them writable in NUT. I have no idea why the writable
variables always also get the ST_FLAG_STRING flag - Arnaud, do you
know if this is necessary / sensible? Perhaps it is just a
copy-and-paste error.

Alfred, do you know if there are any other potentially writable
variables? Potential candidates are:

battery.charge.warning
input.transfer.low
input.transfer.high

"ups.beeper.status" is also writable in principle, but this is already
taken care of by instant commands.

Regarding the other change - NUT's status flags don't have slots for
"RemainingTimeLimitExpired" and "BatteryPresent". So adding these has
no effect other than removing some debug messages. 

-- Peter


Alfred Ganz wrote:
> 
> Gentlemen,
> 
> I have now had a chance to run some initial tests of my changes to
> apc-hid.c of the current development tree. Running newhipups and 
> inspecting the -DDDD output seems to indicate that with the fix below
> everything seems to work as expected. However, I have not yet shut
> my system in order to take the UPS out of action so that I can do my
> various tests with a lightbulb load. I still hope to do this before the 
> end of this week.
> 
> In the initial tests I found that a trick that I was using with the 2.0.2 
> sources no longer works. I have therefore made a change for the second 
> item in my previous message:
> >2) the rather lengthy, unsolicited status messages from my "Back-UPS ES 650"
> >   return two variables that are not contained in the current table,
> >   adding them avoids the indications that newhidups doesn't know what
> >   to do with the variables, although the current change doesn't affect
> >   status evaluation in any way either.
> Rather than using a NULL item mapping I am now using the hex_conversion 
> mapping to accept the two items:
> 	UPS.PowerSummary.PresentStatus.BatteryPresent and
> 	UPS.PowerSummary.PresentStatus.RemainingTimeLimitExpired
> that are however still ignored, as it happened with the 2.0.2 changes. 
> 
> If the experts think it would be worthwhile to actually make the two 
> items available (although they are not particularly useful), please let 
> me know how they might be mapped, and I will add and test such a mapping.
> 
> Here follows the modified patch, AG
> 
> ---------------------------------------------------------------------------
> diff -ru nut-2.1.dev.orig/drivers/apc-hid.c nut-2.1.dev/drivers/apc-hid.c
> --- nut-2.1.dev.orig/drivers/apc-hid.c	2005-09-24 00:37:25.000000000 -0400
> +++ nut-2.1.dev/drivers/apc-hid.c	2005-10-02 14:38:00.000000000 -0400
> @@ -115,11 +115,14 @@
>    
>    /* Battery page */
>    { "battery.charge", 0, 1, "UPS.PowerSummary.RemainingCapacity", NULL, "%.0f", HU_FLAG_OK, NULL },
> -  { "battery.charge.low", ST_FLAG_STRING, 5, "UPS.PowerSummary.RemainingCapacityLimit", NULL,
> -    "%.0f", HU_FLAG_OK, NULL }, /* Read only */
> +  /* the following is writeable for the Back-UPS ES 650 */
> +  { "battery.charge.low", ST_FLAG_RW | ST_FLAG_STRING, 5, "UPS.PowerSummary.RemainingCapacityLimit", NULL,
> +    "%.0f", HU_FLAG_OK | HU_FLAG_SEMI_STATIC, NULL },
>    { "battery.charge.warning", ST_FLAG_STRING, 5, "UPS.PowerSummary.WarningCapacityLimit", NULL, "%.0f", HU_FLAG_OK, NULL }, /* Read only */
>    { "battery.runtime", 0, 0, "UPS.PowerSummary.RunTimeToEmpty", NULL, "%.0f", HU_FLAG_OK, NULL },
> -  { "battery.runtime.low", 0, 0, "UPS.Battery.RemainingTimeLimit", NULL, "%.0f", HU_FLAG_OK, NULL },
> +  /* the following is writeable for the Back-UPS ES 650 */
> +  { "battery.runtime.low", ST_FLAG_RW | ST_FLAG_STRING, 5, "UPS.PowerSummary.RemainingTimeLimit", NULL,
> +    "%.0f", HU_FLAG_OK | HU_FLAG_SEMI_STATIC, NULL },
>    { "battery.voltage",  0, 0, "UPS.PowerSummary.Voltage", NULL, "%.1f", HU_FLAG_OK, NULL },
>    { "battery.voltage.nominal", 0, 0, "UPS.Battery.ConfigVoltage", NULL,
>      "%.1f", HU_FLAG_OK, NULL },
> @@ -161,6 +164,11 @@
>      "%.0f", HU_FLAG_OK, &overbatt_info[0] },
>    { "ups.status", 0, 1, "UPS.PowerSummary.PresentStatus.NeedReplacement", NULL,
>      "%.0f", HU_FLAG_OK, &replacebatt_info[0] },
> +  /* The following two are returned with long notification from Back-UPS ES 650 */
> +  { "ups.status", 0, 1, "UPS.PowerSummary.PresentStatus.RemainingTimeLimitExpired", NULL,
> +    "%.0f", HU_FLAG_OK, &hex_conversion[0]},
> +  { "ups.status", 0, 1, "UPS.PowerSummary.PresentStatus.BatteryPresent", NULL,
> +    "%.0f", HU_FLAG_OK, &hex_conversion[0]},
>  
>    /* Input page */
>    { "input.voltage", 0, 0, "UPS.Input.Voltage", NULL, "%.1f", HU_FLAG_OK, NULL },
> ---------------------------------------------------------------------------
> 
> -- 
>  ----------------------------------------------------------------------
>    Alfred Ganz					alfred-ganz at agci.com
>    AG Consulting, Inc.				(203) 624-9667
>    440 Prospect Street # 11
>    New Haven, CT 06511
>  ----------------------------------------------------------------------
> 
> _______________________________________________
> Nut-upsdev mailing list
> Nut-upsdev at lists.alioth.debian.org
> http://lists.alioth.debian.org/mailman/listinfo/nut-upsdev
>