[mrtg] MRTG vs. Cricket

[Patrick Topping]

Dan,

Makes sense.  I may set up the extra CFG to compare.  I did not even 
think about the difference.....  Thanks for pointing that out.   Have a 
great weekend.

-Patrick

--------------------------------------

McDonald, Dan wrote:
> On Fri, 2009-01-30 at 15:05 -0800, Patrick Topping wrote:
>   
>> I have been polling the 64-bit counters from the beginning.  Below is a 
>> sample from the CFG file:
>>
>> Target[xxx.xxx.xxx_TenGigabitEthernet10_4]: 
>> \TenGigabitEthernet10/4:XXXXX at XXXXX:::::2
>>
>> I know that the graphs are off because I still have Cricket graphing the 
>> same devices as MRTG.  I left Cricket in place while I migrated 
>> everything over to MRTG.  When I look at the graphs on the same 10-gig 
>> interface and look at the actual 10-gig interface on the router, the 
>> Cricket graphs seem be more accurate.
>>     
>
> so, you are polling the locIfOutBitsSec and locIfInBitsSec variables
> with Cricket?  Those would match the "show interface" values on the
> router, but they are not accurate.  They are exponentially weighted to
> give preference to the last 15 seconds of traffic, rather than a
> straight-line average of the number of bytes actually transmitted.
>
> The ifHCInOctets and ifHCOutOctets will give you statistically better
> information over the long term.  And precise numbers is you want totals.
>
> If you want to compare, just create a separate target for the locIf{In|
> Out}BitsSec OIDs, set options[]: guage, and see if those match
> precisely.
>
>
>   
>> I have changed the cron from running every 5 minutes to running every 
>> minute.  I will see how well this does and update the group from there.  
>> Thanks for all the help so far.
>>
>> -Patrick
>>
>> ------------------------------------
>>
>> Sean Cheesman wrote:
>>     
>>> Are you polling v2?  v2 uses 64 bit counters, versus 32 bit v1, which 
>>> isn't enough for anything over like 600Mbps (I don't recall the actual 
>>> rollover point).
>>> ------------------------------------------------------------------------
>>> *From:* mrtg-bounces at lists.oetiker.ch [mrtg-bounces at lists.oetiker.ch] 
>>> On Behalf Of Patrick Topping [ptopping at pobox.com]
>>> *Sent:* Friday, January 30, 2009 12:33 PM
>>> *To:* 'mrtg at lists.oetiker.ch'
>>> *Subject:* [mrtg] MRTG vs. Cricket
>>>
>>> I am hoping someone can help me out with this issue I am having.....  I
>>> am working on migrating from Cricket to MRTG.  So far so good except
>>> that the Cricket graphs are pretty spot on for a 5 minute interval while
>>> the MRTG graphs are anywhere from 600 Mbps to 1 Gbps off at peak time.
>>> For example, I have a 10-gig circuit that runs roughly at 7 Gbps a peak
>>> time.  Cricket graphs it damn near spot on.  MRTG is off by almost a
>>> full gig.
>>>
>>> I have checked the cron jobs and both services are running every 5
>>> minutes.  They are synced to NTP and the clocks are not off so they are
>>> both seeing the exact same data.  I have set up my CFG files to use the
>>> 64 bit counters.
>>>
>>> I have Googled for answers but not a lot out there on this topic.  I am
>>> hoping this is something small I have missed and can easily be fixed.
>>> Any ideas?  Thanks in advance.
>>>
>>> -Patrick
>>>
>>>
>>> _______________________________________________
>>> mrtg mailing list
>>> mrtg at lists.oetiker.ch
>>> https://lists.oetiker.ch/cgi-bin/listinfo/mrtg
>>>       
>> _______________________________________________
>> mrtg mailing list
>> mrtg at lists.oetiker.ch
>> https://lists.oetiker.ch/cgi-bin/listinfo/mrtg
>>     
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> mrtg mailing list
>> mrtg at lists.oetiker.ch
>> https://lists.oetiker.ch/cgi-bin/listinfo/mrtg
>>