Generic SNMP poller via spine getting data but not graphing

[proc]

I've got a new Cacti install that I am experimenting with as a replacement for Cricket and DRRAW, and I have hit a snag with my first service using the "Generic OID Template".

I've configured it to grab from .1.3.6.1.4.1.2021.52.101.1, but the graph itself isn't showing any data. All of the other canned gathering components (interface stats, disk stats, etc) work fine.

Turning on debugging and monitoring the poller (once I moved to spine, I was actually able to get some decent debug output) showed:

Code: Select all

08/27/2008 03:20:02 PM - SPINE: Poller[0] Host[2] DS[20] SNMP: v1: myhost.mydomain.net, dsname: snmp_oid, oid: .1.3.6.1.4.1.2021.52.101.1, value: 287

and further along I see:

Code: Select all

INSERT INTO poller_output (local_data_id, rrd_name, time, output) VALUES (8,'cpu_nice','2008-08-27 15:20:02','23871'),(9,'cpu_system','2008-08-27 15:20:02','24026393'),(10,'cpu_user','2008-08-27 15:20:02','14642524'),(11,'load_1min','2008-08-27 15:20:02','0.17'),(12,'load_15min','2008-08-27 15:20:02','0.21'),(13,'load_5min','2008-08-27 15:20:02','0.18'),(14,'mem_buffers','2008-08-27 15:20:02','317872'),(15,'mem_cache','2008-08-27 15:20:02','5558024'),(16,'mem_free','2008-08-27 15:20:02','5473048'),(17,'traffic_out','2008-08-27 15:20:02','783400470'),(17,'traffic_in','2008-08-27 15:20:02','1921543197'),(18,'hdd_used','2008-08-27 15:20:02','45907872'),(18,'hdd_free','2008-08-27 15:20:02','16608456'),(19,'hdd_used','2008-08-27 15:20:02','60526260'),(19,'hdd_free','2008-08-27 15:20:02','206401720'),(20,'snmp_oid','2008-08-27 15:20:02','287')

And finally:

Code: Select all

08/27/2008 03:20:03 PM - POLLER: Poller[0] CACTI2RRD: /usr/bin/rrdtool update /var/www/html/cacti.mydomain.net/cacti-0.8.7b/rra/test_snmp_oid_20.rrd --template snmp_oid 1219864802:287

However, the "poller_output" table doesn't have anything in it and the rrd file looks like:

Code: Select all

rrdtool dump medical_manager_snmp_oid_20.rrd | more
<!-- Round Robin Database Dump --><rrd> <version> 0003 </version>
        <step> 300 </step> <!-- Seconds -->
        <lastupdate> 1219866001 </lastupdate> <!-- 2008-08-27 15:40:01 EDT -->

        <ds>
                <name> snmp_oid </name>
                <type> GAUGE </type>
                <minimal_heartbeat> 600 </minimal_heartbeat>
                <min> 0.0000000000e+00 </min>
                <max> 1.0000000000e+02 </max>

                <!-- PDP Status -->
                <last_ds> 287 </last_ds>
                <value> NaN </value>
                <unknown_sec> 1 </unknown_sec>
        </ds>

<!-- Round Robin Archives -->   <rra>
                <cf> AVERAGE </cf>
                <pdp_per_row> 1 </pdp_per_row> <!-- 300 seconds -->

                <params>
                <xff> 5.0000000000e-01 </xff>
                </params>
                <cdp_prep>
                        <ds>
                        <primary_value> NaN </primary_value>
                        <secondary_value> NaN </secondary_value>
                        <value> NaN </value>
                        <unknown_datapoints> 0 </unknown_datapoints>
                        </ds>
                </cdp_prep>
                <database>
                        <!-- 2008-08-25 13:45:00 EDT / 1219686300 --> <row><v> NaN </v></row>
                        <!-- 2008-08-25 13:50:00 EDT / 1219686600 --> <row><v> NaN </v></row>
                        <!-- 2008-08-25 13:55:00 EDT / 1219686900 --> <row><v> NaN </v></row>
...

So I know it is getting the data, and it looks like it is putting it in to the RRD file once (overwriting the "last_ds" value each time), but that's it.. the rest is getting dropped to the floor. And the sql query looks right, but it isn't getting entered.

I've checked the sql query, and that (as the cacti user) works. At this point, I'm thinking I am missing a step or something along the way here..

[gandalf]

Please see debugging help at 2nd link of my sig
Reinhard

[proc]

I did manage to solve this by doing the following:
o Delete the graph entirely
o Delete the host (probably not necessary)
o Add the host
o Click the link for "Create Graphs for this Host"
o Follow the prompts/pages

The problem seems to be that if I create a graph via the "Graph Management" page, I get this problem consistently.

However, if I click on the "Devices" link, then select a device/host, and add the graph that way, it works fine.

I even went so far as to blow away the entire cacti install (since it was a new install), and re-do the entire cacti install (including the DB), install the plugin arch, and npc.