Howdy Tom,

So first, I went ahead and did a get against the two OIDs to see what they report. First the AP9606:

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-1 upsAdvOutputLoad.0

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 71

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-1 upsHighPrecOutputLoad

Error in packet

Reason: (noSuchName) There is no such variable name in this MIB.

Failed object: PowerNet-MIB::upsHighPrecOutputLoad

Second the AP9630:

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-2 upsAdvOutputLoad.0

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-2 upsHighPrecOutputLoad.0

PowerNet-MIB::upsHighPrecOutputLoad.0 = Gauge32: 728

With regard to the OID being used on the install, this is interesting. for the AP9606 unit, it's:

/opt/observium/rrd/isp-su5k-1.mydomain.com/sensor-capacity-apc-upsAdvOutputL oad.0.rrd

And for the AP9630, it's:

/opt/observium/rrd/isp-su5k-2.mydomain.com/sensor-capacity-apc-upsHighPrecOu tputLoad.0.rrd

So obviously ya'll did good in figuring out in the discovery process as to which OID should be used. :D

Interestingly, if I'm following the thought correctly, if you're using the upsHighPrecOutputLoad.0 result and dividing that by 10 to get the actual percentage, but if you're taking the result of either upsHighPrecOutputLoad.0 *OR* upsAdvOutputLoad.0 and then dividing by 10, that could be what is throwing the graphs off by that factor of 10. I just don't know PHP enough as to figure out where to look for this.

(and yes, obviously, I changed my snmp community string and domain name before copy/pasting.) ;-)

.Ron

From: observium [mailto:observium-bounces@observium.org] On Behalf Of Tom Laermans Sent: Tuesday, November 25, 2014 10:05 AM To: Observium Network Observation System Subject: Re: [Observium] APC UPS Current is zero

Hi Ron,

Thanks for the somewhat longer reply than mine. ;-)

I have some devices with 9631 cards (which are 9630 with UIO sockets as far as I know). We don't use upsAdvOutputLoad on those, but upsHighPrecOutputLoad. (At least on my 1- and 3-phase devices) - can you check the value returned? And perhaps which OID we are using on your install (quickly via the graph -> rrd command -> check the filename used, or alternatively via the 'sensors' table in the database).

If the 9606 would not report this OID at all, we'd ignore it, but 0 is a valid measurement unfortunately; especially as I know APC tends to show 0 until the actual value is >=1 due to accuracy reasons.

Tom

On 11/25/2014 03:49 PM, Ron Marosko wrote:

Hi Tomas,

This is an issue with the older AP960x management cards, firmware, and APC MIBs. I have two SmartUPS 5000 units, one is an older -RM model which has the AP9606 management card, and one is a newer -RMT5U model with an AP9630 management card. Both of these units are running a Cisco 7206VXR chassis, a Cisco 7609 chassis, and a Dell PowerEdge 1955 chassis, all with dual (or quad in PE1955) power supplies equally balanced between the two units. The UPS with the AP9606 reports 0A of current, and a battery output load of 7.10%, whilst the UPS with the AP9630 reports 13.8A of current, and a battery output load of 71.5%. Even though the power configuration is exactly the same on both sides. Take a wild guess which one I tend to believe is more accurate. ;-)

Just for comparison, walking the PowerNet-MIB on both units yields the following:

AP9606:

PowerNet-MIB::upsBasicIdentModel.0 = STRING: "SMART-UPS 5000 RM"

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

PowerNet-MIB::upsAdvOutputCurrent.0 = Gauge32: 0

AP9630:

PowerNet-MIB::upsBasicIdentModel.0 = STRING: "Smart-UPS 5000"

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

PowerNet-MIB::upsAdvOutputCurrent.0 = Gauge32: 14

Note that last entry. the AP9606 does not report a value for output current.(upsAdvOutputCurrent.0 = Gauge32: 0) Ooops.

One thing I *do* find more interesting is that the output load number (upsAdvOutputLoad.0 = Gauge32: 72) is actually correct, but for some reason on the device page, the output load is showing as being divided by 10. See attached image for that.

Dunno if we need to open a bug on that or what. Thoughts from our esteemed authors on this?

.Ron

From: observium [mailto:observium-bounces@observium.org] On Behalf Of chott@praha1.net Sent: Tuesday, November 25, 2014 5:57 AM To: observium@observium.org Subject: [Observium] APC UPS Current is zero

Hello,

We have monitored APC Smart-UPS 1500 RM with module AP9605

But Capacity/Output load is 11% and Current Output is 0A

Tomas

_______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

Hi Ron,

Scaling of the sensors is defined per OID, so they don't interact. Furthermore, multiplier for the upsAdvOutputLoad is set to 1 in our code, so I'm a bit confused why Observium would be dividing it by 10 for you.

Can you check the value of sensor_multiplier for that sensor in the sensors_table?

Thanks, Tom

On 25/11/2014 18:51, Ron Marosko wrote:

Apologies for self-quote, discovered I typo'd that second snmpget on the AP9606, forgot to add the ".0" on the end of the OID name. The actual result, however, is the same:

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-1 upsHighPrecOutputLoad.0

Error in packet

Reason: (noSuchName) There is no such variable name in this MIB.

Failed object: PowerNet-MIB::upsHighPrecOutputLoad.0

...Ron

*From:*observium [mailto:observium-bounces@observium.org] *On Behalf Of *Ron Marosko *Sent:* Tuesday, November 25, 2014 11:46 AM *To:* 'Observium Network Observation System' *Subject:* Re: [Observium] APC UPS Current is zero

Howdy Tom,

So first, I went ahead and did a get against the two OIDs to see what they report. First the AP9606:

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-1 upsAdvOutputLoad.0

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 71

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-1 upsHighPrecOutputLoad

Error in packet

Reason: (noSuchName) There is no such variable name in this MIB.

Failed object: PowerNet-MIB::upsHighPrecOutputLoad

Second the AP9630:

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-2 upsAdvOutputLoad.0

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m +PowerNet-MIB isp-su5k-2 upsHighPrecOutputLoad.0

PowerNet-MIB::upsHighPrecOutputLoad.0 = Gauge32: 728

With regard to the OID being used on the install, this is interesting... for the AP9606 unit, it's:

/opt/observium/rrd/isp-su5k-1.mydomain.com/sensor-capacity-apc-upsAdvOutputLoad.0.rrd

And for the AP9630, it's:

/opt/observium/rrd/isp-su5k-2.mydomain.com/sensor-capacity-apc-upsHighPrecOutputLoad.0.rrd

So obviously ya'll did good in figuring out in the discovery process as to which OID should be used. :D

Interestingly, if I'm following the thought correctly, if you're using the upsHighPrecOutputLoad.0 result and dividing that by 10 to get the actual percentage, but if you're taking the result of either upsHighPrecOutputLoad.0 **OR** upsAdvOutputLoad.0 and then dividing by 10, that could be what is throwing the graphs off by that factor of 10. I just don't know PHP enough as to figure out where to look for this.

(and yes, obviously, I changed my snmp community string and domain name before copy/pasting.) ;-)

...Ron

*From:*observium [mailto:observium-bounces@observium.org] *On Behalf Of *Tom Laermans *Sent:* Tuesday, November 25, 2014 10:05 AM *To:* Observium Network Observation System *Subject:* Re: [Observium] APC UPS Current is zero

Hi Ron,

Thanks for the somewhat longer reply than mine. ;-)

I have some devices with 9631 cards (which are 9630 with UIO sockets as far as I know). We don't use upsAdvOutputLoad on those, but upsHighPrecOutputLoad. (At least on my 1- and 3-phase devices) - can you check the value returned? And perhaps which OID we are using on your install (quickly via the graph -> rrd command -> check the filename used, or alternatively via the 'sensors' table in the database).

If the 9606 would not report this OID at all, we'd ignore it, but 0 is a valid measurement unfortunately; especially as I know APC tends to show 0 until the actual value is >=1 due to accuracy reasons.

Tom

On 11/25/2014 03:49 PM, Ron Marosko wrote:

Hi Tomas,

This is an issue with the older AP960x management cards, firmware,
and APC MIBs. I have two SmartUPS 5000 units, one is an older --RM
model which has the AP9606 management card, and one is a newer
--RMT5U model with an AP9630 management card. Both of these units
are running a Cisco 7206VXR chassis, a Cisco 7609 chassis, and a
Dell PowerEdge 1955 chassis, all with dual (or quad in PE1955)
power supplies equally balanced between the two units.  The UPS
with the AP9606 reports 0A of current, and a battery output load
of 7.10%, whilst the UPS with the AP9630 reports 13.8A of current,
and a battery output load of 71.5%.  Even though the power
configuration is exactly the same on both sides. Take a wild guess
which one I tend to believe is more accurate. ;-)

Just for comparison, walking the PowerNet-MIB on both units yields
the following:

AP9606:

PowerNet-MIB::upsBasicIdentModel.0 = STRING: "SMART-UPS 5000 RM"

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

PowerNet-MIB::upsAdvOutputCurrent.0 = Gauge32: 0

AP9630:

PowerNet-MIB::upsBasicIdentModel.0 = STRING: "Smart-UPS 5000"

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

PowerNet-MIB::upsAdvOutputCurrent.0 = Gauge32: 14

Note that last entry... the AP9606 does not report a value for
output current.(upsAdvOutputCurrent.0 = Gauge32: 0) Ooops.

One thing I **do** find more interesting is that the output load
number (upsAdvOutputLoad.0 = Gauge32: 72) is actually correct, but
for some reason on the device page, the output load is showing as
being divided by 10. See attached image for that.

Dunno if we need to open a bug on that or what. Thoughts from our
esteemed authors on this?

...Ron

*From:*observium [mailto:observium-bounces@observium.org] *On
Behalf Of *chott@praha1.net <mailto:chott@praha1.net>
*Sent:* Tuesday, November 25, 2014 5:57 AM
*To:* observium@observium.org <mailto:observium@observium.org>
*Subject:* [Observium] APC UPS Current is zero

Hello,

We have monitored APC Smart-UPS 1500 RM with module AP9605

But Capacity/Output load is 11% and Current Output is 0A

Tomas



_______________________________________________

observium mailing list

observium@observium.org  <mailto:observium@observium.org>

http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

---

observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

Login to your Observium database and run:

SELECT `sensor_multiplier` FROM `sensors` WHERE `sensor_id` = <sensor_id>;

Regards,

SG

On 11/26/2014 9:51 AM, Ron Marosko wrote:

Howdy Tom,

Ok you’re hitting my limit of server-side knowledge. How do I do this: “Can you check the value of sensor_multiplier for that sensor in the sensors_table?”

Thanks,

 Ron

*From:*observium [mailto:observium-bounces@observium.org] *On Behalf Of *Tom Laermans *Sent:* Tuesday, November 25, 2014 5:48 PM *To:* Observium Network Observation System *Subject:* Re: [Observium] APC UPS Current is zero

Hi Ron,

Scaling of the sensors is defined per OID, so they don't interact. Furthermore, multiplier for the upsAdvOutputLoad is set to 1 in our code, so I'm a bit confused why Observium would be dividing it by 10 for you.

Can you check the value of sensor_multiplier for that sensor in the sensors_table?

Thanks, Tom

On 25/11/2014 18:51, Ron Marosko wrote:

Apologies for self-quote, discovered I typo’d that second snmpget
on the AP9606, forgot to add the “.0” on the end of the OID name.
The actual result, however, is the same:

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m
+PowerNet-MIB isp-su5k-1 upsHighPrecOutputLoad.0

Error in packet

Reason: (noSuchName) There is no such variable name in this MIB.

Failed object: PowerNet-MIB::upsHighPrecOutputLoad.0

…Ron

*From:*observium [mailto:observium-bounces@observium.org] *On
Behalf Of *Ron Marosko
*Sent:* Tuesday, November 25, 2014 11:46 AM
*To:* 'Observium Network Observation System'
*Subject:* Re: [Observium] APC UPS Current is zero

Howdy Tom,

So first, I went ahead and did a get against the two OIDs to see
what they report. First  the AP9606:

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m
+PowerNet-MIB isp-su5k-1 upsAdvOutputLoad.0

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 71

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m
+PowerNet-MIB isp-su5k-1 upsHighPrecOutputLoad

Error in packet

Reason: (noSuchName) There is no such variable name in this MIB.

Failed object: PowerNet-MIB::upsHighPrecOutputLoad

Second the AP9630:

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m
+PowerNet-MIB isp-su5k-2 upsAdvOutputLoad.0

PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

root@observium:/opt/observium/mibs# snmpget -v 1 -c readstring -m
+PowerNet-MIB isp-su5k-2 upsHighPrecOutputLoad.0

PowerNet-MIB::upsHighPrecOutputLoad.0 = Gauge32: 728

With regard to the OID being used on the install, this is
interesting… for the AP9606 unit, it’s:

/opt/observium/rrd/isp-su5k-1.mydomain.com/sensor-capacity-apc-upsAdvOutputLoad.0.rrd

And for the AP9630, it’s:

/opt/observium/rrd/isp-su5k-2.mydomain.com/sensor-capacity-apc-upsHighPrecOutputLoad.0.rrd

So obviously ya’ll did good in figuring out in the discovery
process as to which OID should be used. :D

Interestingly, if I’m following the thought correctly, if you’re
using the upsHighPrecOutputLoad.0 result and dividing that by 10
to get the actual percentage, but if you’re taking the result of
either upsHighPrecOutputLoad.0 **OR** upsAdvOutputLoad.0 and then
dividing by 10, that could be what is throwing the graphs off by
that factor of 10. I just don’t know PHP enough as to figure out
where to look for this.

(and yes, obviously, I changed my snmp community string and domain
name before copy/pasting.) ;-)

…Ron

*From:*observium [mailto:observium-bounces@observium.org] *On
Behalf Of *Tom Laermans
*Sent:* Tuesday, November 25, 2014 10:05 AM
*To:* Observium Network Observation System
*Subject:* Re: [Observium] APC UPS Current is zero

Hi Ron,

Thanks for the somewhat longer reply than mine. ;-)

I have some devices with 9631 cards (which are 9630 with UIO
sockets as far as I know). We don't use upsAdvOutputLoad on those,
but upsHighPrecOutputLoad. (At least on my 1- and 3-phase devices)
- can you check the value returned? And perhaps which OID we are
using on your install (quickly via the graph -> rrd command ->
check the filename used, or alternatively via the 'sensors' table
in the database).

If the 9606 would not report this OID at all, we'd ignore it, but
0 is a valid measurement unfortunately; especially as I know APC
tends to show 0 until the actual value is >=1 due to accuracy reasons.

Tom

On 11/25/2014 03:49 PM, Ron Marosko wrote:

    Hi Tomas,

    This is an issue with the older AP960x management cards,
    firmware, and APC MIBs. I have two SmartUPS 5000 units, one is
    an older –RM model which has the AP9606 management card, and
    one is a newer –RMT5U model with an AP9630 management card.
    Both of these units are running a Cisco 7206VXR chassis, a
    Cisco 7609 chassis, and a Dell PowerEdge 1955 chassis, all
    with dual (or quad in PE1955) power supplies equally balanced
    between the two units.  The UPS with the AP9606 reports 0A of
    current, and a battery output load of 7.10%, whilst the UPS
    with the AP9630 reports 13.8A of current, and a battery output
    load of 71.5%.  Even though the power configuration is exactly
    the same on both sides. Take a wild guess which one I tend to
    believe is more accurate. ;-)

    Just for comparison, walking the PowerNet-MIB on both units
    yields the following:

    AP9606:

    PowerNet-MIB::upsBasicIdentModel.0 = STRING: "SMART-UPS 5000 RM"

    PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

    PowerNet-MIB::upsAdvOutputCurrent.0 = Gauge32: 0

    AP9630:

    PowerNet-MIB::upsBasicIdentModel.0 = STRING: "Smart-UPS 5000"

    PowerNet-MIB::upsAdvOutputLoad.0 = Gauge32: 72

    PowerNet-MIB::upsAdvOutputCurrent.0 = Gauge32: 14

    Note that last entry… the AP9606 does not report a value for
    output current.(upsAdvOutputCurrent.0 = Gauge32: 0) Ooops.

    One thing I **do** find more interesting is that the output
    load number (upsAdvOutputLoad.0 = Gauge32: 72) is actually
    correct, but for some reason on the device page, the output
    load is showing as being divided by 10. See attached image for
    that.

    Dunno if we need to open a bug on that or what. Thoughts from
    our esteemed authors on this?

    …Ron

    *From:*observium [mailto:observium-bounces@observium.org] *On
    Behalf Of *chott@praha1.net <mailto:chott@praha1.net>
    *Sent:* Tuesday, November 25, 2014 5:57 AM
    *To:* observium@observium.org <mailto:observium@observium.org>
    *Subject:* [Observium] APC UPS Current is zero

    Hello,

    We have monitored APC Smart-UPS 1500 RM with module AP9605

    But Capacity/Output load is 11% and Current Output is 0A

    Tomas




    _______________________________________________

    observium mailing list

    observium@observium.org  <mailto:observium@observium.org>

    http://postman.memetic.org/cgi-bin/mailman/listinfo/observium




_______________________________________________

observium mailing list

observium@observium.org  <mailto:observium@observium.org>

http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

---

observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium

------ Original Message ------ From: "Tom Laermans" tom.laermans@powersource.cx To: "Observium Network Observation System" observium@observium.org Sent: 11/25/2014 10:05:04 AM Subject: Re: [Observium] APC UPS Current is zero

Hi Ron,

Thanks for the somewhat longer reply than mine. ;-)

I have some devices with 9631 cards (which are 9630 with UIO sockets as far as I know). We don't use upsAdvOutputLoad on those, but upsHighPrecOutputLoad. (At least on my 1- and 3-phase devices) - can you check the value returned? And perhaps which OID we are using on your install (quickly via the graph -> rrd command -> check the filename used, or alternatively via the 'sensors' table in the database).

Seems I should add those "data dump" pages for more entity types... :)

adam.

Yep… as long as I stay at or below layer 4 I’m ok… but you get me into this higher layer stuff and I get lost fairly fast. ;-)

From: observium [mailto:observium-bounces@observium.org] On Behalf Of Adam Armstrong Sent: Wednesday, November 26, 2014 12:09 PM To: Observium Network Observation System Subject: Re: [Observium] APC UPS Current is zero

------ Original Message ------

From: "Tom Laermans" tom.laermans@powersource.cx

To: "Observium Network Observation System" observium@observium.org

Sent: 11/25/2014 10:05:04 AM

Subject: Re: [Observium] APC UPS Current is zero

Hi Ron,

Thanks for the somewhat longer reply than mine. ;-)

I have some devices with 9631 cards (which are 9630 with UIO sockets as far as I know). We don't use upsAdvOutputLoad on those, but upsHighPrecOutputLoad. (At least on my 1- and 3-phase devices) - can you check the value returned? And perhaps which OID we are using on your install (quickly via the graph -> rrd command -> check the filename used, or alternatively via the 'sensors' table in the database).

Seems I should add those "data dump" pages for more entity types... :)

adam.