[Nut-upsuser] HP R12000/3 UPS reports status as OL DISCHRG OB

[Charles Lepple]

Nice, that MIB fills in a few details.

.1.3.6.1.4.1.232.165 is cpqPower.

.1.3.6.1.4.1.232.165.3 is cpqPower.ups.

On Jul 16, 2013, at 10:25 PM, Charles Lepple wrote:

> Here's the first update of ups.status:
> 
>   0.102449	getting data: ups.status (.1.3.6.1.4.1.232.165.3.4.5.0)
>   0.102546	su_ups_get: ups.status .1.3.6.1.4.1.232.165.3.4.5.0
>   0.102646	nut_snmp_get(.1.3.6.1.4.1.232.165.3.4.5.0)
>   0.104539	SNMP UPS driver : entering su_status_set()
>   0.104641	su_find_infoval: found OL (value: 3)
>   0.104730	=> value: 3
> 
>   0.104823	getting data: ups.status (.1.3.6.1.4.1.232.165.3.2.5.0)
>   0.104919	su_ups_get: ups.status .1.3.6.1.4.1.232.165.3.2.5.0
>   0.105028	nut_snmp_get(.1.3.6.1.4.1.232.165.3.2.5.0)
>   0.107134	SNMP UPS driver : entering su_status_set()
>   0.107231	su_find_infoval: found DISCHRG (value: 2)
>   0.107319	=> value: 2
> 
>   0.107412	getting data: ups.status (.1.3.6.1.4.1.232.165.3.7.3.0)
>   0.107507	su_ups_get: ups.status .1.3.6.1.4.1.232.165.3.7.3.0
>   0.107616	nut_snmp_get(.1.3.6.1.4.1.232.165.3.7.3.0)
>   0.109529	SNMP UPS driver : entering su_status_set()
>   0.109630	su_find_infoval: found OB (value: 1)
>   0.109720	=> value: 1
> 
> The first one (...3.4.5.0) is labeled in the driver as CPQPOWER_OID_POWER_STATUS. I checked around for this, and didn't find a good definition, but that seems to match the OL state.

Confirmed by the MIB:

upsOutputSource OBJECT-TYPE
    SYNTAX  INTEGER {
    other(1),
    none(2),
    normal(3),				-- normal, single UPS module output
    bypass(4),
    battery(5),
    booster(6),				-- Single or Double Boost, line-interactive UPSs only
    reducer(7),				-- Buck, line-interactive UPSs only
    parallelCapacity(8),		-- normal enhanced by Parallel for Capacity operation
    parallelRedundant(9),		-- normal enhanced by Redundant Parallel operation
    highEfficiencyMode(10)		-- normal enhanced by High Efficiency mode
    }
    ACCESS  read-only
    STATUS  mandatory
    DESCRIPTION
	   "The present source of output power.  The enumeration
	       none(2) indicates that there is no source of output
	       power (and therefore no output power), for example,
	       the system has opened the output breaker."
    ::= { upsOutput  5 }

> The second one seems specific to the battery:
> 
> static info_lkp_t cpqpower_battery_abm_status[] = {
>        { 1, "CHRG" },
>        { 2, "DISCHRG" },
> /*      { 3, "Floating" }, */
> /*      { 4, "Resting" }, */
> /*      { 5, "Unknown" }, */
>        { 0, "NULL" }
> } ;
> 
> I could see the battery monitoring circuit discharging the battery occasionally for maintenance, but is there a separate front-panel indicator for this mode?

The MIB suggests that the "discharge" state is part of normal battery management:

upsBatteryAbmStatus OBJECT-TYPE
    SYNTAX  INTEGER {
    batteryCharging(1),
    batteryDischarging(2),
    batteryFloating(3),
    batteryResting(4),
    unknown(5)
	}
    ACCESS  read-only
    STATUS  mandatory
    DESCRIPTION
	"Gives the status of the Advanced Battery Management;
	 batteryFloating(3) status means that the charger is temporarily 
	 charging the battery to its float voltage; batteryResting(4) is the 
	 state when the battery is fully charged and none of the other actions 
	 (charging/discharging/floating) is being done."
    ::= { upsBattery 5 }

> The third one is CPQPOWER_OID_ALARM_OB. This definitely doesn't make sense, since things with "alarm" in the name are usually pretty direct.


I looked up the third OID, and it does not seem to map to that alarm status. Instead, it is "upsTestTrap" and while the MIB calls it read-write, the only defined action is for writing a startTestTrap to start a test.

Arnaud: the CPQPOWER_OID_ALARM_OB entry looks like it was a guess [*]. Do you know of any units which require this? Otherwise, I think we should take it out.

[*] https://github.com/networkupstools/nut/commit/c1b83e7d40083263719de13e631a72dfc96961a9

-- 
Charles Lepple
clepple at gmail