Cisco specific VPDN Traffic Statistics for L2TP Tunnels

[gandalf]

I've taken the opportunity to create this template for a networking company in Michigan. It is based on Cisco VPDN MIB and runs at least for L2TP Tunnels.
It will graph the L2TP Tunnel connections. Graphs are exactly the same as the standard traffic graphs, so no screenshots attached. If tunnels are reconnected, cacti re-associates traffic to previous tunnel connections using the same radius login credentials (= Session User Name). This requires re-index method to be set to Verify All Fields!
To get an idea, the table is shown that appears when "Create Graphs for this Host" is selected.
Reinhard

[yeled]

[quote="gandalf"]
It will graph the L2TP Tunnel connections. Graphs are exactly the same as the standard traffic graphs, so no screenshots attached. If tunnels are reconnected, cacti re-associates traffic to previous tunnel connections using the same radius login credentials (= Session User Name). This requires re-index method to be set to [b]Verify All Fields[/b]!
[/quote]

I cannot get this working? it asks for hostname, snmp version, community etc - fine. but it also asks for "Index Type", "Index Value" & "Output Type ID" when I create the graphs??

[yeled]

it was missing the data query section. pretty simple to setup. thanks!

create this Data Query called "L2TP" (or whathaveyou)

"<path_cacti>/resource/snmp_queries/cisco_vpdn_session.xml" is the line that scrolls off.

[nevet5m]

I've been having problems getting responses from my device. I can walk it and get the relevant ids for each session in the tunnel but when it comes to getting info (in this case Bytes Out) for a specific session the router complains:

Code: Select all

[root@jsycacti ~]# snmpwalk -On -v2c -cxxxxx 10.252.20.3 1.3.6.1.4.1.9.10.24.1.3.2.1.6
.1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13351 = Counter32: 62661756 bytes
.1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13352 = Counter32: 62239068 bytes
.1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13353 = Counter32: 62239570 bytes
.1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13354 = Counter32: 1082244277 bytes
.1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13355 = Counter32: 62243486 bytes
.1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13356 = Counter32: 62243120 bytes
[root@jsycacti ~]#
[root@jsycacti ~]# snmpget -On -v2c -cxxxxx 10.252.20.3 1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13351
.1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13351 = No Such Instance currently exists at this OID
[root@jsycacti ~]#

Does anyone know what would cause this? Cheers.

[shd]

I've been having problems getting responses from my device. I can walk it and get the relevant ids for each session in the tunnel but when it comes to getting info (in this case Bytes Out) for a specific session the router complains:

Code: Select all

[root@jsycacti ~]# snmpget -On -v2c -cxxxxx 10.252.20.3 1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13351
.1.3.6.1.4.1.9.10.24.1.3.2.1.6.2.48975.13351 = No Such Instance currently exists at this OID
[root@jsycacti ~]#

Does anyone know what would cause this? Cheers.

/necroposting mode on
This problem appears on some old IOS versions, such as 12.4.
15.1 replies to snmpget without problem.
/necroposting mode off

[Lithium]

I've taken the opportunity to create this template for a networking company in Michigan. It is based on Cisco VPDN MIB and runs at least for L2TP Tunnels.
It will graph the L2TP Tunnel connections. Graphs are exactly the same as the standard traffic graphs, so no screenshots attached. If tunnels are reconnected, cacti re-associates traffic to previous tunnel connections using the same radius login credentials (= Session User Name). This requires re-index method to be set to Verify All Fields!
To get an idea, the table is shown that appears when "Create Graphs for this Host" is selected.
Reinhard

nice templates

[somasundaram]

Hi,

i am trying to poll the Virtual-Access & Virtual-Template interfaces from a 7206 router using this scripts. i am able to discover the interface after hitting "create graphs for this Host". But i am missing something that the graph's are not plotted.

here is the log file.

06/13/2014 02:40:07 PM - SPINE: Poller[0] Host[228] TH[1] Recache DataQuery[1] OID: .1.3.6.1.2.1.1.3.0, output: 1585894841
06/13/2014 02:40:07 PM - SPINE: Poller[0] Host[228] TH[1] Recache DataQuery[16] OID: .1.3.6.1.2.1.1.3.0, output: 1585894841
06/13/2014 02:40:07 PM - SPINE: Poller[0] Host[228] TH[1] Total Time: 0.062 Seconds

Below is the graph debug output.

RRDTool Command:
C:/rrdtool/rrdtool.exe graph - \
--imgformat=PNG \
--start=-86400 \
--end=-300 \
--title="If:|query_cvpdnSessionInterfaceName| - |query_cvpdnSessionAttrUserName| - Traffic (Total Bandwidth)" \
--base=1000 \
--height=120 \
--width=570 \
--alt-autoscale-max \
--lower-limit="0" \
--vertical-label="bits per second" \
--slope-mode \
--font TITLE:8: \
--font AXIS:7: \
--font LEGEND:8: \
--font UNIT:7: \
DEF:a="C\:/Apache2/htdocs/cacti/rra/228/6498.rrd":"BytesIn":AVERAGE \
DEF:b="C\:/Apache2/htdocs/cacti/rra/228/6498.rrd":"BytesOut":AVERAGE \
CDEF:cdefa="a,8,*" \
CDEF:cdeff="b,8,*" \
AREA:cdefa#00CF00FF:"Inbound" \
GPRINT:cdefa:LAST:" Current\:%8.2lf %s" \
GPRINT:cdefa:AVERAGE:"Average\:%8.2lf %s" \
GPRINT:cdefa:MAX:"Maximum\:%8.2lf %s\n" \
COMMENT:"Total In\: 0 bytes\n" \
LINE1:cdeff#002A97FF:"Outbound" \
GPRINT:cdeff:LAST:"Current\:%8.2lf %s" \
GPRINT:cdeff:AVERAGE:"Average\:%8.2lf %s" \
GPRINT:cdeff:MAX:"Maximum\:%8.2lf %s\n" \
COMMENT:"Total Out\: 0 bytes\n"


Can some one guide me what i'm missing here.

Regards
Soma

[somasundaram]

Hi,

i did a mistake that, i have not selected the re-index method as "verify all fields', but even after modifying the graphs are not getting plotted.

SNMP: v2: dsname: traffic_in, oid: .1.3.6.1.2.1.31.1.1.1.6.184, value: 3370378874
SNMP: v2: dsname: traffic_out, oid: .1.3.6.1.2.1.31.1.1.1.10.184, value: 961982190

Here are the snapshots of my config

Graph Template

graph template.jpg (160 KiB) Viewed 6976 times

Data Template

data template.jpg (118.04 KiB) Viewed 6976 times

Data Query

data query.jpg (86.99 KiB) Viewed 6976 times

Gandalf - can you help me to over come this.


Regards
Soma

[somasundaram]

Hi,

i did a mistake that, i have not selected the re-index method as "verify all fields', but even after modifying the graphs are not getting plotted.

SNMP: v2: dsname: traffic_in, oid: .1.3.6.1.2.1.31.1.1.1.6.184, value: 3370378874
SNMP: v2: dsname: traffic_out, oid: .1.3.6.1.2.1.31.1.1.1.10.184, value: 961982190

Here are the snapshots of my config

Graph Template

graph template.jpg

Data Template

data template.jpg

Data Query

data query.jpg

Gandalf - can you help me to over come this.


Regards
Soma

Now It is working, i am able to get the usage graphs. Just re-imported the templates & added the data queries that's it.

But whenever one vpdn user session alone gets reconnects, it is getting a different virtual-access interface. how the new virtual-access interface will get associated to the existing username for that the graph has been created. For my case i have reset a session and the virtual interface ID got changed from Vi160 to Vi441.

But the graph already created for the same username "xxx" was still reflects the old interface id which is Vi160.

Dear Reinhard - need your expertise to fix this. can you pl help.


Regards
Soma

[somasundaram]

Hi Gandalf,

This is working now, all what i have changed is to make the re-index method as "index count changed".

Now it is associating the user session into the previous graphs even it reconnects.

But one bottle neck / limitation which i have seen is the maximum no of session or graphs it was able to poll is only 60.

Even for this i the snmp timeout value for that particular host need to set as 2000ms.

when i am trying to add more no of sessions to poll, i am getting the error messages as follows.

12/14/2014 10:40:11 PM - PCOMMAND: Poller[0] Host[433] WARNING: Recache Event Detected for Host
12/14/2014 10:40:10 PM - SYSTEM THOLD STATS: Time:0.0259 Tholds:1 DownHosts:0
12/14/2014 10:40:10 PM - SYSTEM STATS: Time:7.8456 Method:spine Processes:3 Threads:15 Hosts:336 HostsPerProcess:112 DataSources:11898 RRDsProcessed:5965
12/14/2014 10:40:10 PM - SPINE: Poller[0] Host[433] TH[3] DS[8003] WARNING: SNMP timeout detected [4000 ms], ignoring host 'x.x.x.x'
12/14/2014 10:40:09 PM - SPINE: Poller[0] Host[433] TH[3] DS[8003] WARNING: SNMP timeout detected [4000 ms], ignoring host 'x.x.x.x'
12/14/2014 10:40:09 PM - SPINE: Poller[0] Host[433] TH[3] DS[7996] WARNING: SNMP timeout detected [4000 ms], ignoring host 'x.x.x.x'
12/14/2014 10:40:09 PM - SPINE: Poller[0] Host[433] TH[3] DS[7997] WARNING: SNMP timeout detected [4000 ms], ignoring host 'x.x.x.x'
12/14/2014 10:40:09 PM - SPINE: Poller[0] Host[433] TH[3] DS[7996] WARNING: SNMP timeout detected [4000 ms], ignoring host 'x.x.x.x'

But using verbose query method all the out put is displayed by in max 3 seconds. (Just to test i have increased the snmp timeout value into 4000ms but no luck)

can you through some lights in this and help me out please.

Regards
Soma

Cacti-0.8.8a|OS-windows-2008|NET-SNMP- 5.6.1.1|RRDTool-1.4.x|PHP-5.3.17|SPINE-0.8.8a|Hosts-336|Graphs-5951| Polling time-7.7 Seconds|

[mokhtar]

Hello everyone
how can i create the graph

help