The graphs are stacked. Most of the colours actually are slighly transparent, that's how we make the graphs look normal on both light and dark backgrounds.

In general we don't have elements of graphs "overlapping". If multiple things are being drawn in the same domain we usually use lines, when it's something being aggregated we just stack them, as in these graphs.

adam.

Christopher Barry via observium wrote on 27/09/2023 15:24:
Hi all, first post:
Pardon me for jumping in - I likely have no idea what I'm talking about, but that has never stopped me from opening mouth/inserting foot in the past!

It _seems_ like if the solid colors were in fact transluscent, then you could readily see and tell where they were overlapped, and having two overlaid colors and thus interfaces's data occupying the same space would be more readily - wait for it... - _observable_.

Anyway, that popped into my head so I thought I would blurt it out.
I'll go back to lurking now - sorry

Cheers,
-C


On Wed, Sep 27, 2023 at 10:10 AM Adam Armstrong via observium <observium@lists.observium.org <mailto:observium@lists.observium.org>> wrote:

    Same thing, the colours match (though RRD's legend line is so thin
    it's hard to see now, it used to be a diagnonal line which was
    more visible)



    I might switch back to using only the area indicator for legend
    colours, but this graph isn't really intended to be all that
    readable, it's useless for this purpose on most large devices:



    adam.


    Tony Guadagno via observium wrote on 27/09/2023 15:04:

    Adam, yes, I see that, it’s on the transmit side that it does not
    make sense to me.  So receive (above 0) for gig0/0/0 is lt-green,
    but transmit(below 0) for gig0/0/0 is what?? Lt-purple?  How
    would I know that? In this particular case, the transmit and
    receive numbers are roughly equal…but if the traffic was more
    asymmetric…then how would I know that?

    Tony Guadagno

    O +1 585 577 1003

    C +1 585 703 6700

    E to...@guadagnoconsulting.com <mailto:to...@guadagnoconsulting.com>

    cid:image001.jpg@01D84DD6.FC9912E0

    *From:*Adam Armstrong via observium
    <observium@lists.observium.org>
    <mailto:observium@lists.observium.org>
    *Sent:* Wednesday, September 27, 2023 9:58 AM
    *To:* Observium <observium@lists.observium.org>
    <mailto:observium@lists.observium.org>
    *Cc:* Adam Armstrong <ad...@observium.org>
    <mailto:ad...@observium.org>
    *Subject:* [Observium] Re: sorry if this is a dumb question, but
    how do i read traffic graphs???

    The colours do actually match!



    Though these graphs aren't really intended for you to be able to
    pick out individual ports, there are usually too many for that to
    be possible.

    It looks like there's a bug in the colour generating code which
    is generating one too few colours and the last one is wrapping
    back to the first colour. The code is supposed to generate
    $entries number of colours between two points to make a gradient.

    adam.

    Tony Guadagno wrote on 27/09/2023 14:52:

        Adam, I think I am starting to understand, however, here is a
        snippet of the total traffic graph since 6am and another of
        gig0/0/0 same time period:

        Total:                               gig0/0/0

        On the total graph, the transmits and receives from gig0/0/0
        are lt-purple and lt-green.  But in the legend for the total
        traffic, those colors don’t match…I guess that is what I am
        confused about, how do I know (on the total traffic graph)
        which interface is contributing if the color on the graph
        does not match the color on the legend?



        Tony Guadagno

        O +1 585 577 1003

        C +1 585 703 6700

        E to...@guadagnoconsulting.com
        <mailto:to...@guadagnoconsulting.com>

        cid:image001.jpg@01D84DD6.FC9912E0

        *From:*Adam Armstrong via observium
        <observium@lists.observium.org>
        <mailto:observium@lists.observium.org>
        *Sent:* Wednesday, September 27, 2023 8:56 AM
        *To:* Tony Guadagno via observium
        <observium@lists.observium.org>
        <mailto:observium@lists.observium.org>
        *Cc:* Adam Armstrong <ad...@observium.org>
        <mailto:ad...@observium.org>
        *Subject:* [Observium] Re: sorry if this is a dumb question,
        but how do i read traffic graphs???

        Hi Tony,

        This is a stacked graph for a single device's total traffic,
        the layers are stacked in the order they're presented on the
        legend. The legend shows the "out" colours as lines because
        they're "fake" drawn as lines to get positive values in the
        legend.

        Gi0 and Gi0/0/2 in on this system have no traffic, so aren't
        really relevant as they aren't visible on the graph. It's a
        little odd the colours have cycled with only 4 entries,
        that's likely just a quirk of the colour deciding function.

        Anything positive on the axis is in/receive, and anything
        negative is out/transmit, the same as our other graphs.

        adam.

        Tony Guadagno via observium wrote on 27/09/2023 13:45:


            Hi, after several years of using Observium, it just
            occurred to me recently that I am not sure how to read
            the traffic graphs.  For example, I always thought that
            above “0” on the horizontal axis was transmit and below
            was receive.  But as I look closer, that cannot be. 
            Also, the colors on the graph do not seem to line up with
            the legend…for example in this graph, I see 4 colors,
            light/dark green and light/dark purple.  But the legend
            seems to have the same color for g0/0/0 as g0.

            I guess my questions are:

            How are transmits and receives represented on the traffic
            graphs?

            Why are the colors in the graph not lining up with the
            legend?

            What does negative 100M represent?

            thanks



            Tony Guadagno

            O +1 585 577 1003

            C +1 585 703 6700

            E to...@guadagnoconsulting.com
            <mailto:to...@guadagnoconsulting.com>

            cid:image001.jpg@01D84DD6.FC9912E0





            _______________________________________________

            observium mailing list --observium@lists.observium.org
            <mailto:observium@lists.observium.org>

            To unsubscribe send an email toobservium-le...@lists.observium.org
            <mailto:observium-le...@lists.observium.org>



    _______________________________________________
    observium mailing list --observium@lists.observium.org 
<mailto:observium@lists.observium.org>
    To unsubscribe send an email toobservium-le...@lists.observium.org
    <mailto:observium-le...@lists.observium.org>

    _______________________________________________
    observium mailing list -- observium@lists.observium.org
    <mailto:observium@lists.observium.org>
    To unsubscribe send an email to
    observium-le...@lists.observium.org
    <mailto:observium-le...@lists.observium.org>



_______________________________________________
observium mailing list -- observium@lists.observium.org
To unsubscribe send an email to observium-le...@lists.observium.org

_______________________________________________
observium mailing list -- observium@lists.observium.org
To unsubscribe send an email to observium-le...@lists.observium.org

Reply via email to