i've created a new snmp data query, that has two references in the index:
snmpwalk ---- 1.3.6.1.4.1.10923.1.1.1.1.7.1.11.1.2
SNMPv2-SMI::enterprises.10923.1.1.1.1.7.1.11.1.2.26.5 = STRING: "PCRF-gy"
SNMPv2-SMI::enterprises.10923.1.1.1.1.7.1.11.1.2.31.1 = STRING: "Apollo-OCS"
.....
so the index is 26.5, or 31.1, etc...
here's the data query xml (incomplete):
Code: Select all
<interface>
<name>CC-AS Statistics</name>
<description>listing the service-based charging (FBC) CC-AS statistics for each APN.</description>
<oid_index>.1.3.6.1.4.1.10923.1.1.1.1.7.1.11.1.2</oid_index>
<oid_index_parse>OID/REGEXP:^.{37}(.*)</oid_index_parse>
<index_order>ccasIndex</index_order>
<index_order_type>alphabetic</index_order_type>
<fields>
<ccasIndex>
<name>Index</name>
<source>index</source>
<direction>input</direction>
</ccasIndex>
<ccasName>
<name>CC AS Name</name>
<method>walk</method>
<source>value</source>
<direction>input</direction>
<oid>.1.3.6.1.4.1.10923.1.1.1.1.7.1.11.1.2</oid>
</ccasName>
<apnName>
<name>APN Name</name>
<method>walk</method>
<source>OID/REGEXP:^.{38}(.*)\.[0-9]{1,3}$</source>
<direction>input</direction>
<oid>.1.3.6.1.4.1.10923.1.1.1.1.7.1.11.1.2</oid>
</apnName>
<ccasStartReq>
<name>Start Requests</name>
<method>walk</method>
<source>value</source>
<direction>output</direction>
<oid>.1.3.6.1.4.1.10923.1.1.1.1.7.1.11.1.3</oid>
</ccasStartReq>
the query looks fine in verbose mode in the host view,
data source looks ok too, with the index values populated,
graphs are created - but the rrd file does not exist.
if i switch back to the cmd.php poller, the rdd gets created and updates fine.
switching back to spine - the rrd is no longer updated....
running spine from command line for the host yields (incomplete):
02/02/2011 05:05:29 PM - SPINE: Poller[0] Host[25] TH[1] RECACHE: Processing 13 items in the auto reindex cache for '10.32.10.9'
No log handling enabled - turning on stderr logging
snmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failure02/02/2011 05:05:29 PM - SPINE: Poller[0] Host[25] TH[1] NOTE: There are '555' Polling Items for this Host
02/02/2011 05:05:29 PM - SPINE: Poller[0] Host[25] TH[1] DEBUG: HOST COMPLETE: About to Exit Host Polling Thread Function
graphs for other hosts dont suffer with spine - so this is not an across the board issue with snmp
cmd.php show same errors, but it doesn't seen to affect it:
# php -q /var/www/cacti/cmd.php 25 25
No log handling enabled - turning on stderr logging
snmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failuresnmp_build: unknown failure
it seems like there is something spine doesn't like about that indexed query in particular, that cmd.php copes with...
this is the only multi index data query i have, so maybe the regexps aren't right?
i tried the debugging walk through, cache clearing, etc...
both spine and cacti are patched and running 0.8.7.g
SNMP Version NET-SNMP version: 5.3.2.2
thank you in advance for your help