[smokeping-users] Smoke cut-off problem back in 2.4.0?

[Dan Hamilton]

Hello everyone,

I'm glad this discussion as resurfaced, as I've been meaning to ask 
about it again, I just haven't had time.

But I think we really have two different discussions going on here:

1) issues with graph scaling
2) issue of averaging of historical detail

Regarding issue #2: Ged and Stephen, I see where you're coming from, but 
I also understand Tobi's reasoning as well.

I don't know much about the inner workings of RRDtool, but would it be 
possible to have the option of storing peak values rather than average 
values in the historical data? (I mean the peak of the median values, 
not the peak of the smoke.)  This might be more appealing for some people.

I haven't played with the email alerts yet myself, but that would help 
as well in providing notification of historic problems.

Regarding issue #1:  I would love to have some config options to change 
the way the graph scaling works:

1) the current method, with an arbitrary multiplier (ie, 1.5x median)
2) scaled to accommodate the highest smoke value
3) fixed scaling to an arbitrary value

This way, everyone can display their data in the way that makes the most 
sense for them.

Regards,
Dan

Stephen Spence wrote:
> Indeed, this has proven to be a big issue for us also.
> 
> We use smokeping to monitor the quality of our various branch office links and unless I grab current graphs of network events it's very hard to identify issues and provide meaningful information later on.
> 
> Is there some switch to disable the reduction of detail? Storage and processing capacity really isn't an problem for us here.
> 
> 
> -----Original Message-----
> From: smokeping-users-bounces at lists.oetiker.ch [mailto:smokeping-users-bounces at lists.oetiker.ch] On Behalf Of G.W. Haywood
> Sent: 12 June 2008 13:23
> To: Tobias Oetiker
> Cc: Smokeping users' mailing list
> Subject: Re: [smokeping-users] Smoke cut-off problem back in 2.4.0?
> 
> Hi Tobi,
> 
> On Thu, 12 Jun 2008, Tobias Oetiker wrote:
> 
>> you can configure rrdtool to keep hiresolution data for arbitrary
>> time. The consolidation is just a way to keep more data with less
>> space. If you have the space, configure the hires RRAs to be longer
>> (still have the low res rras around for long term graphs).
>> Performance will not be impacted in any way by this change.
> 
> I don't think I'm making my point clearly enough.
> 
> I know that that RRDtool, and the Terabyte of storage on my desk,
> permit me to keep records at high resolution for extended periods.
> 
> However, short of painstakingly examining the high-resolution graphs
> minute-by-minute I do not see how Smokeping will alert me to transient
> short duration events.  These may be symptoms of some unknown problem
> which appeared while I was away on holiday: It's 8am on Monday morning
> and there's a great big heap of paper in my 'IN' tray.  An ugly crowd
> of disgruntled computer users is gathering at the door to the IT dep't
> offices, and I'd like to take a quick look at what's been going on in
> the LAN for the past month before I unlock it - so that I don't appear
> to be quite so clueless as is in fact the case.
> 
> Networks can be a bit like combat - long periods of nothing happening
> punctuated by moments of sheer terror.  The 'nothing happening' bits
> establish reference points but little else, and can be characterized
> with little storage (or screen) capacity.  The punctuations are more
> interesting, and will require more resources.  At the moment these
> punctuations tend to be masked.  In your words:
> 
>> Date: Sat, 26 Apr 2008 14:49:38 +0200 (CEST)
>> From: Tobias Oetiker <tobi at oetiker.ch>
>> To: G.W. Haywood <ged at jubileegroup.co.uk>
>> In-Reply-To: <Pine.LNX.4.58.0804261259001.18067 at mail3.jubileegroup.co.uk>
>> Cc: smokeping-users at lists.oetiker.ch
>> Subject: Re: [smokeping-users] Graph scale problem
>>
>> [snip] the net result is, that the spike becomes lower as it gets
>> distributed across time [snip]
> 
> That's the bit that bothers me.
> 
> I think I've banged on this drum enough now. :)
> 
> --
> 
> 73,
> Ged.
>