BIND 9 Statistics for Cacti

[jbeardsley]

where i can download the tar.gz,,,,,

link : http://uversaconsulting.net/download/bi ... 1.0.tar.gz ----- not working" =(

I found a copy here:

http://files.pcplus.ge/Public/Linux/Mon ... 1.0.tar.gz

[JJX]

After reading lot of topics, i am confused...

- Is there a version working with BIND 9.7.2-P3 ?

- http://www.bayour.com/bind9-snmp/
Should I use version 1.7 or 1.pre8? I dont see any changelog any the files

- My BIND is running to a differencet machine from my cacti installation.

* bind9-snmp-stats.pl plus BayourCOM_SNMP.pm & BAYOUR-COM-MIB.txt should be installed to DNS server?

* pass <OID> /usr/bin/perl path_to/bind9-stats-snmpd.pl

Do I need to run this command? Which OID is the script searching to the snmp tree?

Thank you

[JJX]

The solution was this post:
http://forums.cacti.net/viewtopic.php?p=177247#p177247 [bind9-stats-2.0.tar ]

After using this script I almost solved the graphing problems...
The only problem is that snmpwalk reutns only 2 zones (GLOBAL And an empty one) although names.stats exports data for all zones...

Any idea?

thank you

[smstone]

The solution was this post:
http://forums.cacti.net/viewtopic.php?p=177247#p177247 [bind9-stats-2.0.tar ]

After using this script I almost solved the graphing problems...
The only problem is that snmpwalk reutns only 2 zones (GLOBAL And an empty one) although names.stats exports data for all zones...

Any idea?

thank you

Hello JJX,

Not sure if you're still having this problem, but I was using the same version as you and having the same problem as well (only GLOBAL zone showing). I realized the bind9-stats-snmpd.pl was trying to parse my Bind 9.7.1_p2 version named.stats file, which has a new format so it was only reporting GLOBAL and not adding the needed OIDs for the rest of my zones. In this thread, JeffRoberson's updated bind9-stats-snmpd.pl here: http://forums.cacti.net/about3018-165.html worked for me and now is showing all the zones in Cacti.

Hope this helps!

-smstone

[vanpeer]

I can use the OID in snmpd.conf :
pass .1.3.6.1.4.1.2021.55 /usr/bin/perl /usr/local/bin/bind9-stats-snmpd.pl

because this entry .1.3.6.1.4.1.2021.55 is used by another template :

exec .1.3.6.1.4.1.2021.55 hdIndex /usr/local/bin/snmpdiskio hdIndex
Can i use another OID for Bind-stat ?

Thanks for your answer

[someone17]

The solution was this post:
http://forums.cacti.net/viewtopic.php?p=177247#p177247 [bind9-stats-2.0.tar ]

Hi,

I have graphing problem with this script because half of the time are graphs empty, but statistics is correctly displayed.

On other host I have a lot of zones and script doesn't work (it's too slow).

Any suggestions? Tnx.

[j0ca]

Hello,

I am using the bind9-stats-1.0, the SNMP configuration, it doesn't give any problem if I use the command snmpbulkwalk or snmpwalk.

But when I use the Cacti Data Query (BIND 9 Statistics (SNMP) ) it returns the following result:

Code: Select all

+ Executing SNMP walk for list of indexes @ '.1.3.6.1.4.1.2021.55.1' Index Count: 0
+ No SNMP data returned

I used wireshark to analyse the SNMP requests and I noticed that the BIND 9 Statistics (SNMP) makes 2 getbulkrequest for the same OID.

I believe the problem is in Cacti poller, since the shell (snmp) commands work correctly.

I am using a Cacti 0.8.7h clean install on a Centos 5.7 box.

Can someone help?

[albech]

Hello,

I am using the bind9-stats-1.0, the SNMP configuration, it doesn't give any problem if I use the command snmpbulkwalk or snmpwalk.

But when I use the Cacti Data Query (BIND 9 Statistics (SNMP) ) it returns the following result:

Code: Select all

+ Executing SNMP walk for list of indexes @ '.1.3.6.1.4.1.2021.55.1' Index Count: 0
+ No SNMP data returned

I used wireshark to analyse the SNMP requests and I noticed that the BIND 9 Statistics (SNMP) makes 2 getbulkrequest for the same OID.

I believe the problem is in Cacti poller, since the shell (snmp) commands work correctly.

I am using a Cacti 0.8.7h clean install on a Centos 5.7 box.

Can someone help?

I increased the SNMP Timeout for the particular host being polled and that solved this problem.

[CommandanteFerdl]

Hi!

We are using bind9-stats-2.0.tar here, together with BIND 9.6, net-snmp-5.7.1 and cacti-0.8.7e4.
The graph this script produces looks fine, except for some occasions where there are unusual spikes
which I cannot really explain. Does anybody else experience this behaviour? It seems to just go away
after a while, but it's a bit annoying. Screenshot attached ...

[ptselios]

Does anybody have a solution for the 64bit systems?
I have some busy caching DNS servers that I "monitor" with Cacti. The problem is that the servers are 64bit and the Cacti is setup in 32bit.
That creates a lot of issues because I need to "reset" my counters every 2 days. Is there anyway to enable 64bit counters for the script of Cory Powers (bind9-stats-snmpd.pl)?

[ofi]

Hi!

We are using bind9-stats-2.0.tar here, together with BIND 9.6, net-snmp-5.7.1 and cacti-0.8.7e4.
The graph this script produces looks fine, except for some occasions where there are unusual spikes
which I cannot really explain. Does anybody else experience this behaviour? It seems to just go away
after a while, but it's a bit annoying. Screenshot attached ...

IMHO race-condition. Cacti poller has 5min interval and so does bind-stats cron. I`m currently facing the same issue.