Bluecoat ( Cacheflow) proxy OID's

[Morgan]

Here are the usefull OID's to use and graph for a bluecoat (ex cacheflow) proxy

please snmpget these from a command line to make sure you use proper max values. mine might defer from yours based on usage etc.

CPU
.1.3.6.1.4.1.3417.2.4.1.1.1.4.1

use gauge for data source

Objects in cache
.1.3.6.1.3.25.17.3.1.2.0

use gauge (make sure your max value is high enough)

Server HTTP Errors
.1.3.6.1.3.25.17.3.2.2.2.0
COUNTER
This corresponds to queries from the proxy to internet http servers that returned an http error code (ex 404 page not found etc)

Server HTTP Requests
.1.3.6.1.3.25.17.3.2.2.1.0
COUNTER
This corresponds to queries from the proxy to internet http servers

Server HTTP Traffic
Inbound: .1.3.6.1.3.25.17.3.2.2.3.0
Outbound: .1.3.6.1.3.25.17.3.2.2.4.0
Value is in KBITS as per the mib, so remember to use *1024 cdef.
COUNTER

Inbound will be much higher than outbound, so i used area for inbound.
Inbound is HTTP Traffic FROM internet web server to the proxy, outbound is http traffic from the proxy to those servers.

Client HTTP Requests & Hits
Requests OID: .1.3.6.1.3.25.17.3.2.1.1.0
Hits OID: .1.3.6.1.3.25.17.3.2.1.2.0
COUNTER
Requests corresponds to HTTP requests received by the proxy from clients. Hits is how many of those requests were served by objects in the proxy cache.

Client HTTP Errors
.1.3.6.1.3.25.17.3.2.1.3.0
COUNTER
This corresponds to the requests from clients to the proxy that returned an HTTP error.

Client HTTP Traffic
Inbound: .1.3.6.1.3.25.17.3.2.1.4.0
Outbound: .1.3.6.1.3.25.17.3.2.1.5.0
Value is in KBITS as per the mib, so remember to use *1024 cdef.
COUNTER

Outbound will be much higher than inbound here, so i used area for outbound and line 1 for inbound. Outbound represents HTTP traffic to clients from the proxy, inbound represents http traffic from the client to the proxy.

[cguerra]

I got the following problem. I created the SNMP query for proxy according to http://forums.cacti.net/about3595.html but i got the following problem :

For Example: CPU usage

CPU
.1.3.6.1.4.1.3417.2.4.1.1.1.4.1

use gauge for data source


I created the rrd

/usr/bin/rrdtool create \
/var/www/cacti/rra/proxy_1_snmp_oid_8.rrd \
--step 60 \
DS:snmp_oid:GAUGE:120:0:100 \
RRA:AVERAGE:0.5:1:14400 \
RRA:AVERAGE:0.5:5:20160 \
RRA:AVERAGE:0.5:30:14000 \
RRA:AVERAGE:0.5:120:15500 \
RRA:AVERAGE:0.5:288:14400 \
RRA:MAX:0.5:1:14400 \
RRA:MAX:0.5:5:20160 \
RRA:MAX:0.5:30:14000 \
RRA:MAX:0.5:120:15500 \
RRA:MAX:0.5:288:14400 \

But it get the following error on the LOG

07/08/2008 06:29:02 PM - CMDPHP: Poller[0] Host[8] DS[8] WARNING: Result from SNMP not valid. Partial Result:

When i check on a snmpwalk manually i got :

[root@hefesto rra]# snmpwalk -On -c public -v 1 proxy-1 .1.3.6.1.4.1.3417.2.4.1.1.1.4.1
.1.3.6.1.4.1.3417.2.4.1.1.1.4.1 = INTEGER: 2


There is a mismatch on the OID type. it is defined as INTEGER but i created it as GAUGE and i suspect this is the reason why it fails.

Any suggestion?

I got same proble for the Objects in cache, they should be defined as gauge but when i do snmpwalk they reply as Counter32

Objects in cache
.1.3.6.1.3.25.17.3.1.2.0

use gauge (make sure your max value is high enough)

[root@hefesto rra]# snmpwalk -On -c public -v 1 proxy-1 1.3.6.1.3.25.17.3.1.2.0
.1.3.6.1.3.25.17.3.1.2.0 = Counter32: 1298