Nut upsuser - Mar 2014 - snmp-ups sends status "OL OB" on HP R3000 UPS with AF465A management card [UPDATE]

Further information on this issue, attn. Arnaud Quette:

Here is the result of a test I made a few days ago. I've pulled the
input plug on the UPS and watched the 3 critical OIDs through a MIB browser:

- 1.3.6.1.4.1.232.165.3.4.5.0 goes from 3 (OL) to 5 (OB)
  (as expected => good)

- 1.3.6.1.4.1.232.165.3.7.3.0 stays at 1 (OB)
  (=> seems meaningless)

- 1.3.6.1.4.1.232.165.3.7.4.0 doesn't exist
  (either before or after pulling the plug)
  (supposed to carry 'LB' info => not good)

Thanks in advance for any feedback you might have on this issue.

If you need more or different debugging information, please let me know.

TIA

Cheers. Bye.

Ph. A.


On 27/02/14 15:25, Philippe Andersson wrote:
> Additional question: I noticed on the HP Website that the latest version
> of "CPQPOWER.MIB" (1.76, dated Sept. 19, 2013) was more recent
than the
> related file (./drivers/compaq-mib.c) in the NUT driver.
> 
> Is the MIB structure as implemented on the management card perhaps no
> longer in line with the NUT driver ?
> 
> The latest MIB file can be downloaded from the following URL:
> 
>
<https://h20565.www2.hp.com/portal/site/hpsc/template.PAGE/public/psi/swdDetails/?javax.portlet.begCacheTok=com.vignette.cachetoken&javax.portlet.endCacheTok=com.vignette.cachetoken&javax.portlet.prp_bd9b6997fbc7fc515f4cf4626f5c8d01=wsrp-navigationalState%3Didx%3D|swItem%3DMTX_d32e012fd5f14f1fb3e3859897|itemLocale%3D|mode%3D|action%3DdriverDocument&javax.portlet.tpst=bd9b6997fbc7fc515f4cf4626f5c8d01&ac.admitted=1393496865122.876444892.492883150>
> 
> TIA
> 
> Cheers. Bye.
> 
> Ph. A.
> 
> 
> On 27/02/14 11:06, Philippe Andersson wrote:
>> Hello list,
>>
>> I've installed NUT on an HP blade server, and I try to get it
>> communicating via SNMP with 2 HP R3000 UPSes, each fitted with a AF465A
>> management card. To do this, I used the snmp-ups driver, and configured
>> it for the "cpqpower" MIB. Communication with the UPSes is
established.
>>
>> My problem is that both UPSes constantly return the "OL OB"
status
>> which, in addition to being nonsensical, drives upsmon crazy (it flips
>> between the two states every second).
>>
>> Here is the upsc output (from one of the UPSes -- they're both
similar):
>>
>>
-----------------------------<cut>--------------------------------
>> mcbackup:/etc/ups # upsc R3000-B at localhost
>> ambient.temperature: 28.0
>> ambient.temperature.high: 40.00
>> ambient.temperature.low: 0.00
>> battery.charge: 99.00
>> battery.current: 0.00
>> battery.runtime: 7440.00
>> battery.voltage: 13.10
>> device.mfr: HP
>> device.model: HP RT3000 G2 UPS
>> device.serial: 3C83435450
>> device.type: ups
>> driver.name: snmp-ups
>> driver.parameter.mibs: auto
>> driver.parameter.pollfreq: 15
>> driver.parameter.pollinterval: 2
>> driver.parameter.port: mcups1.cal.org
>> driver.parameter.snmp_version: v1
>> driver.version: 2.6.5
>> driver.version.data: cpqpower MIB 1.5
>> driver.version.internal: 0.68
>> input.current: 0.40
>> input.frequency: 50.00
>> input.phases: 1.00
>> input.quality: 2.00
>> input.voltage: 209.00
>> input.voltage.nominal: 230
>> outlet.1.id: 1.00
>> outlet.1.status: on
>> outlet.1.switchable: yes
>> outlet.2.id: 2.00
>> outlet.2.status: on
>> outlet.2.switchable: yes
>> outlet.count: 2.00
>> outlet.desc: All outlets
>> outlet.id: 0
>> output.current: 0.20
>> output.frequency: 50.00
>> output.frequency.nominal: 500
>> output.phases: 1.00
>> output.voltage: 223.00
>> output.voltage.nominal: 230
>> ups.delay.shutdown: 20
>> ups.delay.start: 30
>> ups.firmware: 01.06.0026;01.02.0004
>> ups.load: 19.00
>> ups.mfr: HP
>> ups.model: HP RT3000 G2 UPS
>> ups.realpower: 3000.00
>> ups.serial: 3C83435450
>> ups.status: OL OB             <================ !!!
>> ups.test.result: Unknown
>> ups.timer.shutdown: 0.00
>> ups.timer.start: 0.00
>> ups.type: normal
>>
-----------------------------<cut>--------------------------------
>>
>> If I use a MIB browser to inspect the .1.3.6.1.4.1.232.165.3.4.5.0 OID
>> (defined in ./drivers/compaq-mib.c as
"CPQPOWER_OID_POWER_STATUS"), the
>> value returned by both UPSes is 3, meaning "OL" as defined in
>> "cpqpower_pwr_info[]" (which is indeed the expected state for
the UPS at
>> present.
>>
>> But if I inspect OID .1.3.6.1.4.1.232.165.3.7.3.0
>> ("CPQPOWER_OID_ALARM_OB") as suggested by Arnaud, this one
returns "1"
>> on both UPSes (i.e. "OB", as defined in
"cpqpower_alarm_ob[]") -- an
>> information which is clearly inaccurate. Both UPSes are currently on
>> grid power, as reported by their Web management interface.
>>
>> Both UPS management cards (HP P/N AF465A) are on the latest firmware
>> version (v. 1.01.013).
>>
>> My system:
>> OpenSUSE 13.1 64bit on HP BL460c in a C7000 enclosure
>> nut-2.6.5-6.1.5.x86_64
>> nut-drivers-net-2.6.5-6.1.5.x86_64
>> net-snmp-5.7.2-9.1.2.x86_64
>> (NUT installed from official OpenSUSE repository)
>>
>> UPSes:
>> UPS Name :			HP RT3000 G2 UPS
>> UPS Part Number :		AF468A
>> UPS Firmware Revision :		01.02.0004
>> Communication Board Firmware Revision :	01.06.0026
>> Card Firmware revision :	1.01.013
>> Card Part Number :		AF465A
>> Card Technical Level :		05
>> Card Hardware Revision :	BB
>>
>> Attached is the captured trace of running the driver in debug mode.
>>
>> To me, this look more and more like a bug in the Management card
>> firmware. Has any one of you already experienced this? Is there a
>> workaround?
>>
>> TIA
>>
>> Cheers. Bye.
>>
>> Ph. A.
>>
> 
-- 

*Philippe Andersson*
Unix System Administrator
IBA Particle Therapy |
Tel: +32-10-475.983
Fax: +32-10-487.707
eMail: pan at iba-group.com
<http://www.iba-worldwide.com>


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 263 bytes
Desc: OpenPGP digital signature
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140311/af00802e/attachment.sig>

Hi!

Have a similar issue with one of our customers having HP R12000/3 UPS 
with AF430A card. Upsmon shows the UPS as Online, on Battery and Battery 
discharging at the same time:

[root at smrsrvslv06 lib64]# upsc hpups at localhost
ambient.temperature: 19.0
ambient.temperature.high: 0.00
ambient.temperature.low: 0.00
battery.charge: 94.00
battery.current: 0.00
battery.runtime: 4620.00
battery.voltage: 27.70
device.mfr: Hewlett-Packard
device.model: HP R12000/3UPS
device.serial: P38900C9VY8077
device.type: ups
driver.name: snmp-ups
driver.parameter.pollfreq: 15
driver.parameter.pollinterval: 2
driver.parameter.port: 10.2.1.226
driver.parameter.snmp_version: v1
driver.version: 2.7.1
driver.version.data: cpqpower MIB 1.5
driver.version.internal: 0.70
input.frequency: 50.00
input.L1-N.voltage: 224.00
input.L1.current: 0.70
input.L1.realpower: 0.00
input.L2-N.voltage: 223.00
input.L2.current: 1.00
input.L2.realpower: 0.00
input.L3-N.voltage: 223.00
input.L3.current: 0.60
input.L3.realpower: 0.00
input.phases: 3.00
input.quality: 0.00
outlet.count: 0.00
outlet.desc: All outlets
outlet.id: 0
output.frequency: 50.00
output.L1-N.voltage: 224.00
output.L1.current: 0.60
output.L2-N.voltage: 223.00
output.L2.current: 0.90
output.L3-N.voltage: 225.00
output.L3.current: 0.50
output.phases: 3.00
ups.delay.shutdown: 20
ups.delay.start: 30
ups.firmware:  2.02
ups.L1.realpower: 86.00
ups.L2.realpower: 140.80
ups.L3.realpower: 81.60
ups.load: 25.00
ups.mfr: Hewlett-Packard
ups.model: HP R12000/3UPS
ups.serial: P38900C9VY8077
ups.status: OL DISCHRG OB
ups.test.result: Unknown
ups.timer.shutdown: 0.00
ups.timer.start: 0.00
ups.type: normal
[root at smrsrvslv06 lib64]#
Broadcast message from nut (Mon Mar 31 15:25:57 2014):

UPS hpups at localhost on battery

Actually the UPS is online, battery is charging

SNMP values of some "critical" variables used by NUT snmp-ups driver
to
make an UPS status decision:

enterprises.232.165.3.2.5.0 = INTEGER: 2
Battery status (1-charging, 2-discharging, 3-floating, 4-resting, 5-unknown)
WRONG value, most probably UPS or SNMP card firmware is out of sync with 
HP Power MIB.

enterprises.232.165.3.4.5.0 = INTEGER: 3
Output power source (1-other, 2-none, 3-normal, 4-bypass, 5-battery, 
6-booster, 7-reducer, 8-parallel capacity, 9-parallel redundant, 10-high 
efficiency mode)
Ok, right value

enterprises.232.165.3.7.3.0 = INTEGER: 1
MIB shows this as r/w var to initate/test status of Trap test: "Setting 
startTestTrap to 1 will initiate a TrapTest is sent out from HPMM.  All 
other set values are invalid."
WRONG variable, it is not an "On Battery" flag as is in 
nut-2.7.1/drivers/compaq-mib.c

The variable enterprises.232.165.3.7.4.0 does not exists neither in 
snmpwalk output nor in the MIB
In nut-2.7.1/drivers/compaq-mib.c it is "Low Battery" flag. It seems 
that the MIB does not have a good source to make "Low battery"
decision.

Any ideas?


11.03.2014 14:56, Philippe Andersson wrote:
> Further information on this issue, attn. Arnaud Quette:
>
> Here is the result of a test I made a few days ago. I've pulled the
> input plug on the UPS and watched the 3 critical OIDs through a MIB
browser:
>
> - 1.3.6.1.4.1.232.165.3.4.5.0 goes from 3 (OL) to 5 (OB)
>    (as expected => good)
>
> - 1.3.6.1.4.1.232.165.3.7.3.0 stays at 1 (OB)
>    (=> seems meaningless)
>
> - 1.3.6.1.4.1.232.165.3.7.4.0 doesn't exist
>    (either before or after pulling the plug)
>    (supposed to carry 'LB' info => not good)
--
Ivan Kuznetsov
System Administrator
SOLVO ltd
phone +7 812 6060555
fax   +7 812 6060771
http://www.solvo.ru

Philippe, Ivan,

please see https://github.com/networkupstools/nut/issues/118
and send back your data.

cheers
Arno
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.alioth.debian.org/pipermail/nut-upsuser/attachments/20140405/682f10dd/attachment-0001.html>