1.1.36, XML for data query, RRDtool won't create new files

[charlielamm]

Hi there, I have a working (I think) data query, runs with no errors.
However when I try to create a new graph RRDtool won't create it.

This isn't an issue if I create a script and gather up the data from that.

Can someone please recommend some troubleshooting steps?

In the cacti log, I don't see the poller trying to reach my new device at all.

i can create a new file as cacti user runnign rrdtool from the shell.

[netniV]

Attach the DQ and I can see what's missing if anything.

[charlielamm]

Hi netiv:

Here is the Data Query XML. thanks in advance for your help!

<query>
<name>Get APC netbotz 200 temp Information</name>
<description>Get SNMP based temp information from Netbotz 200</description>
<index_order>SensorName</index_order>

<oid_index>.1.3.6.1.4.1.318.1.1.10.4.2.3.1.3</oid_index>

<oid_num_indexes>.1.3.6.1.4.1.318.1.1.10.4.2.2</oid_num_indexes>

<index_order_type>alphabetic</index_order_type>

<fields>
<SensorName>
<name>SensorName</name>
<method>walk</method>
<source>value</source>
<direction>input</direction>
<oid>.1.3.6.1.4.1.318.1.1.10.4.2.3.1.3</oid>

</SensorName>


<Temp>
<name>Temp</name>
<method>walk</method>
<source>value</source>
<direction>output</direction>
<oid>.1.3.6.1.4.1.318.1.1.10.4.2.3.1.5</oid>
</Temp>

</fields>


</query>

[charlielamm]

BTW one more tidbit about this issue. If I go to data sources > turn on source debug and then copy the "create" debug information into command line on my linux system, a (blank) graph is created. No data yet though

[charlielamm]

This gets stranger. it seems the issue perhaps isn't with the DQ, it may be with the poller....

Check this out:

[root@syslog cacti]# php -q cmd.php --first=10 --last=10
2018/03/14 15:17:52 - POLLER: Poller[1] NOTE: There are no items in your poller for this polling cycle!

Note that my other devices poll just fine--ID=1, ID=5, those are working!

This device, tied to the DQ above, won't graph, and is ID=10 according to the user interface, and yet the Poller says it doesn't exist at all!

I don't know where these ID's are stored/how they work. Anyone know why the webpage would say the device in question is ID10 but the poller seem to think ID10 doesn't exist? Can I change something in the MySQL database somewhere? Does anyone know what table this would live in?

[netniV]

You can turn on Device Debug which will create more debugging information for the host.

Try this MySQL query:

Code: Select all

SELECT * FROM poller_item WHERE id = 10;

[charlielamm]

Thanks. MySQL couldn't find that ID, so I deleted everything and meticulously followed these steps (again)

viewtopic.php?t=11647

Debug is on!

Good news--graphs are now created, all templates etc. can be added in everywhere they need to go.

Bad news--OR GOOD NEWS? For the first time, in the logs, I can see in the logs why nothing is being graphed, but I don't know what to make of it.....

018/03/15 12:15:07 - POLLER: Poller[Main Poller] Device[16] Device[AAA_NETBOTZ200] Graphs[AAA-netbotz200, AAA-netbotz200] DS[AAA-NETBOTZ-200] SNMP: v1: 10.1.100.53, dsname: INTEGER, oid: .1.3.6.1.4.1.318.1.1.10.4.2.3.1.5.1, output: U

2018/03/15 12:20:07 - POLLER: Poller[Main Poller] WARNING: Invalid Response(s), Errors[2] Device[16] Device[AAA_NETBOTZ200] Thread[1] Graphs[AAA-netbotz200, AAA-netbotz200] Graphs[AAA-netbotz200, AAA-netbotz200] DS[AAA-NETBOTZ-200, AAA-NETBOTZ-200]

2018/03/15 12:25:06 - POLLER: Poller[Main Poller] Device[16] Device[AAA_NETBOTZ200] Graphs[AAA-netbotz200, AAA-netbotz200] DS[AAA-NETBOTZ-200] SNMP: v1: 10.1.100.53, dsname: INTEGER, oid: .1.3.6.1.4.1.318.1.1.10.4.2.3.1.5.3, output: U
2018/03/15 12:25:06 - POLLER: Poller[Main Poller] Device[16] Device[AAA_NETBOTZ200] Graphs[AAA-netbotz200, AAA-netbotz200] DS[AAA-NETBOTZ-200] WARNING: Result from SNMP not valid. Partial Result: U

That's not right, the actual "get" is .1.3.6.1.4.1.318.1.1.10.4.2.3.1.5.1.1, so I think an OID suffix is missing.

The index walk from my DQ gets me as far as the being able to see each sensor in the netbotz (there are two: .0 and .1, eg .1.3.6.1.4.1.318.1.1.10.4.2.3.1.5.1 and .1.3.6.1.4.1.318.1.1.10.4.2.3.1.5.0)--that all seems to work

The actual temperatures needed to graph are at these OID: .1.3.6.1.4.1.318.1.1.10.4.2.3.1.5.1.1 and .1.3.6.1.4.1.318.1.1.10.4.2.3.1.5.0.3

Do I need to create 2 DQ's, one for each sensor? I was hoping I could create one for as many sensors as I have hooked up to the netbotz but I can['t figure out how to do this.

[netniV]

1.3.6.1.4.1.318.1.1.10.4.2.2 memSensorsStatusTableSize

This is the number of entries you'll find in

1.3.6.1.4.1.318.1.1.10.4.2.3 memSensorsStatusTable

So, x = 1 to n where n is the number from memSensorsStatusTableSize you will have

1.3.6.1.4.1.318.1.1.10.4.2.3.1.1.x = memSensorsStatusModuleNumber
The number of the module that the sensor is attached to.
1.3.6.1.4.1.318.1.1.10.4.2.3.1.2.x = memSensorsStatusSensorNumber
The number of this sensor on the module.
1.3.6.1.4.1.318.1.1.10.4.2.3.1.3.x = memSensorsStatusSensorName
The sensor name.
1.3.6.1.4.1.318.1.1.10.4.2.3.1.4.x = memSensorsStatusSensorLocation
The location of the sensor.
1.3.6.1.4.1.318.1.1.10.4.2.3.1.5.x = memSensorsTemperature
The sensor's current temperature reading. Units are displayed in the scale shown in the memSensorsStatusSysTempUnits OID
1.3.6.1.4.1.318.1.1.10.4.2.3.1.6.x = memSensorsHumidity
The current humidity reading from the sensor.
1.3.6.1.4.1.318.1.1.10.4.2.3.1.7.x = memSensorsCommStatus
The communications status of the sensor.
1.3.6.1.4.1.318.1.1.10.4.2.3.1.8.x = memSensorsAlarmStatus
The alarm status of the sensor.

Try looking up your OID's on oidref.com as I find most are on there. It also shows brothers and children of the various parts of the OID key you are at so makes finding the correct ones a lot easier. It's where I got the above information from.

[charlielamm]

Nice link!!! Thanks!!!!!

Sadly I don't think that's the problem. X replaced by 1, 2, 3 etc. does not reveal any useful data, for a get or walk.

X has to be replaced by TWO children, not just one, (in my setup anyway) to reveal useful data. For the indexes I presented earlier, or the ones you present in your generous help.

So for instance for temperature X has to be replaced by:

.0.3

which now reveals INTEGER:73

or

.1.1

INTEGER: 74

Same idea for humidity, sensor name etc. X is always equal to 0 or 1, followed by another child.

....and I can't see any way for a DQ to find replacements for X that are two children deep?

Does this make sense? perhaps i am missing something very basic here?

So.....Same problem I state above (although i greatly appreciate your help)

This is a netbotz 200 with a Netbotz 150 "linked" using APCs proprietary sensor linking system/cabling/termination. Not the env monitoring system you would find everywhere--so maybe the OIDs are now literally off the charts? Makes me think--Is this solvable at all w a DQ? It seems to me like maybe not.

[netniV]

If it's that complex, it may be worth switching from a straight DQ to a SS (Script Server Script). That way you can build the OID's the way that you want to without worrying about fitting the DQ basic style.

Unfortunately, I don't have a device like yours so I can't play around and say that the DQ can or can't work specifically for you. But I find it odd that you have to go two children deep as that seems to be at odds with the MIB.