graphing question 🔗 link

I am new to the hobbit and trying to understand how it all works together.

I have my server and client up and running and seems to be doing just fine
but I still do not understand all the nuances of the code so the reason for
this question. 

I have noticed that under the trend the "disk utilization" graphs have four
disk systems plotted on one graph.

Yet under the "disk" icon the "disk utilization" graphs show two graphs. The
first graph has the plot for 3 file systems and the second for one file
systems. 

I assume that these two graphs are created with the same hobbitgraph.cfg. Is
this correct? 

Are these two graphs created differently? 

Thanks

Robert P. McGraw, Jr.
Manager, Computer System                 EMAIL: user-33cf07af04dd@xymon.invalid
Purdue University                         ROOM: MATH-807
Department of Mathematics                PHONE: (XXX) XXX-XXXX
XXX N. University Street                   FAX: (XXX) XXX-XXXX
West Lafayette, IN XXXXX-XXXX

Hi All

 
Quick one. I am running a vbs script called IIS health Check on my
WIndooz machine. It produces a file which is place in the bbwin\tmp
directory and picked up and sent to the XyMon Server. XyMon then
displays the values with a nice graph. 

 
The file looks like this:

<start>

green 2009/02/19 03:18:56 PM

 
IIS Health Check:

&green Connections:     120

&green RequestsQueued:           10

&green Sessions:             20

 
<end>

 
Now I produced a exe which reads performance counter does the same thing
as the IIS check i.e it produces a file called msdiskperf in the
BBwin\tmp directory. However there is no graph. 

The file looks like this:

<start>

green 2009/02/19 01:33:56 PM

 
MS Disk Check:

&green Avg. Disk Queue Length               2.5

&green Avg. Disk Sec/Read         10

&green Avg. Disk Sec/Write        0.1

&green % Disk Time        24

 
<end>

 
What do I need to do to include a graph based on the data? Anyone have
some ideas? The data does get to XyMon there is just no graph.

 
Regards

Neil

From: Neil Franken [mailto:user-1689acfc5a3b@xymon.invalid] Sent: Donnerstag, 19. Februar 2009 14:32
To: user-ae9b8668bcde@xymon.invalid
Subject: [hobbit] Graphing Question

Hi All

Quick one. I am running a vbs script called IIS health Check on my WIndooz machine. It produces a file which is place in the bbwin\tmp directory and picked up and sent to the XyMon Server. XyMon then displays the values with a nice graph. 
The file looks like this:
<start>
green 2009/02/19 03:18:56 PM

IIS Health Check:
&green Connections:     120
&green RequestsQueued:           10
&green Sessions:             20

<end>

Now I produced a exe which reads performance counter does the same thing as the IIS check i.e it produces a file called msdiskperf in the BBwin\tmp directory. However there is no graph. The file looks like this:
<start>
green 2009/02/19 01:33:56 PM

MS Disk Check:
&green Avg. Disk Queue Length               2.5
&green Avg. Disk Sec/Read         10
&green Avg. Disk Sec/Write        0.1
&green % Disk Time        24

<end>

What do I need to do to include a graph based on the data? Anyone have some ideas? The data does get to XyMon there is just no graph.

Regards
Neil

Hi,

do you use Xymon's NCV mechanism to build up and feed the RRD files?

Then you have to use 
NAME : VALUE  
lines in your output; and that's not the case in your performance counter output

Regards
Johann

Hi All

Thanks for the info. My file is being produced in the format Name: Value now but still I cannot see a graph.

My question is the following:
Is the format?
name:{tab}value
Or
Name:Value

Also I use the windows eol indicator is this incorrect?

Regards
Neil

-----Original Message-----
From: Johann Eggers [mailto:user-769b09132207@xymon.invalid] Sent: 19 February 2009 04:20 PM
To: user-ae9b8668bcde@xymon.invalid
Subject: RE: [hobbit] Graphing Question

From: Neil Franken [mailto:user-1689acfc5a3b@xymon.invalid] Sent: Donnerstag, 19. Februar 2009 14:32
To: user-ae9b8668bcde@xymon.invalid
Subject: [hobbit] Graphing Question

Hi All

Quick one. I am running a vbs script called IIS health Check on my WIndooz machine. It produces a file which is place in the bbwin\tmp directory and picked up and sent to the XyMon Server. XyMon then displays the values with a nice graph. 
The file looks like this:
<start>
green 2009/02/19 03:18:56 PM

IIS Health Check:
&green Connections:     120
&green RequestsQueued:           10
&green Sessions:             20

<end>

Now I produced a exe which reads performance counter does the same thing as the IIS check i.e it produces a file called msdiskperf in the BBwin\tmp directory. However there is no graph. The file looks like this:
<start>
green 2009/02/19 01:33:56 PM

MS Disk Check:
&green Avg. Disk Queue Length               2.5
&green Avg. Disk Sec/Read         10
&green Avg. Disk Sec/Write        0.1
&green % Disk Time        24

<end>

What do I need to do to include a graph based on the data? Anyone have some ideas? The data does get to XyMon there is just no graph.

Regards
Neil

Hi,

do you use Xymon's NCV mechanism to build up and feed the RRD files?

Then you have to use 
NAME : VALUE  
lines in your output; and that's not the case in your performance counter output

Regards
Johann

Hi All

 
I am busy helping with some ASP.NET monitoring. I got the performance
counters to report. Now they want graphs so I got a basic graph up and
running. Basically I want to show a graph with two lines. 1 Line
indicates the current requests(green) the other the queued
requests(red). Like I said I got the graph running and I have the two
values reporting. However it appears a little funny. The graph is moving
into the negative range see attached screen shot.

 
What I have seen is that the requests queue are constantly 0. The
current request range from 0-20. Also the number on the y axis reports
it in milli and it is a little misleading. As I am not a rrdtool expert
I think it is averaging the two series and getting this value?

 
Here is my graph definition:

[asphealth]

 
              TITLE ASP Over All Health

              YAXIS Number

              DEF:RequestsQueued=asphealth.rrd:RequestsQueued:AVERAGE

              DEF:RequestCurrent=asphealth.rrd:RequestCurrent:AVERAGE

              LINE2:RequestsQueued#FF0000:RequestsQueued

              LINE2:RequestCurrent#00FF66:RequestCurrent

              COMMENT:\n

              GPRINT:RequestsQueued:LAST:RequestsQueued\: %5.1lf%s (cur)

              GPRINT:RequestsQueued:MAX: \: %5.1lf%s (max)

              GPRINT:RequestsQueued:MIN: \: %5.1lf%s (min)

              GPRINT:RequestsQueued:AVERAGE: \: %5.1lf%s (avg)\n

              GPRINT:RequestCurrent:LAST:RequestCurrent\: %5.1lf%s (cur)

              GPRINT:RequestCurrent:MAX: \: %5.1lf%s (max)

              GPRINT:RequestCurrent:MIN: \: %5.1lf%s (min)

              GPRINT:RequestCurrent:AVERAGE: \: %5.1lf%s (avg)\n

 
A typical data set looks like this:

ASP Health Check:

  ApplicationRestarts:        0

 ApplicationRunning: 3

 RequestsQueued:     0

 RequestCurrent:     6

 
Like I said it is working but the values are misleading. Any of your
bright people got some pointers? I had a read through the rrdtool
documentation and I realised it is a very rich tool. Anyone got some
ideas?

 
Regards

Neil

Hello Neil,

I am busy helping with some ASP.NET monitoring. I got the performance counters to report. Now they want graphs so I got a basic graph up and running. Basically I want to show a graph with two lines. 1 Line indicates the current requests(green) the other the queued requests(red). Like I said I got the graph running and I have the two values reporting. However it appears a little funny. The graph is moving into the negative range see attached screen shot.

The values stored in the RRD are thus unexpected. What is the type of the datasets in the RRD? Perhaps the type is DERIVE while GAUGE is appropriate. That might cause negative values.

HTH,
  Wim Nelis


*******************************************************************************************************
The NLR disclaimer (http://www.nlr.nl/emaildisclaimer) is valid for NLR e-mail messages.
*******************************************************************************************************

Yeah that the problem. I tried doing that and I won myself a stupidity
award. I have two pages which uses the graph. I initially had some of
the values as derive. I deleted one of the asphealth.rrd files after
correcting them. I forgot about the other file and this caused the graph
to look strange.

Regards
Neil

-----Original Message-----
From: W.J.M. Nelis [mailto:user-f4ccfde53c0d@xymon.invalid] Sent: 02 June 2010 09:44 AM
To: user-ae9b8668bcde@xymon.invalid
Subject: Re: [hobbit] Graphing Question

Hello Neil,

The values stored in the RRD are thus unexpected. What is the type of the datasets in the RRD? Perhaps the type is DERIVE while GAUGE is appropriate. That might cause negative values.

HTH,
  Wim Nelis


************************************************************************
*******************************
The NLR disclaimer (http://www.nlr.nl/emaildisclaimer) is valid for NLR
e-mail messages.
************************************************************************
*******************************

One last question.

I need to make the YAxis work on a log scale is there a way I can do
this in the graph definition? Basically I have a lot of 1-100 value in a
series and another series running over 250 000.

Regards
Neil

-----Original Message-----
From: Neil Franken [mailto:user-1689acfc5a3b@xymon.invalid] Sent: 02 June 2010 10:37 AM
To: user-ae9b8668bcde@xymon.invalid
Subject: RE: [hobbit] Graphing Question

Yeah that the problem. I tried doing that and I won myself a stupidity
award. I have two pages which uses the graph. I initially had some of
the values as derive. I deleted one of the asphealth.rrd files after
correcting them. I forgot about the other file and this caused the graph
to look strange.

Regards
Neil

-----Original Message-----
From: W.J.M. Nelis [mailto:user-f4ccfde53c0d@xymon.invalid] Sent: 02 June 2010 09:44 AM
To: user-ae9b8668bcde@xymon.invalid
Subject: Re: [hobbit] Graphing Question

Hello Neil,

The values stored in the RRD are thus unexpected. What is the type of the datasets in the RRD? Perhaps the type is DERIVE while GAUGE is appropriate. That might cause negative values.

HTH,
  Wim Nelis


************************************************************************
*******************************
The NLR disclaimer (http://www.nlr.nl/emaildisclaimer) is valid for NLR
e-mail messages.
************************************************************************
*******************************

Hello Neil.

How about
  -l 1
  -o
in your graph definition?

Regards,

  Wim Nelis.


*******************************************************************************************************
The NLR disclaimer (http://www.nlr.nl/emaildisclaimer) is valid for NLR e-mail messages.
*******************************************************************************************************

Hi fellow XYmonians,
 
I have a script that periodically checks my DSL speed and reports 2 values:
 
Download : 13400801
Upload : 1027604
 
which are Bit/s. Actually it reports MBit/s and I recalculate the Bits.
Now I want to graph these values.
 
This is what I have in graphs.cfg:
 
[speed]
        TITLE DSL Speed Info
        YAXIS Mbit/s
        DEF:Download=speed.rrd:Download:AVERAGE
        DEF:Upload=speed.rrd:Upload:AVERAGE
        LINE2:Download#00CCCC:Download
        LINE2:Upload#FF0000:Upload
        COMMENT:\n
        GPRINT:Download:LAST:Download \: %2.2lf%s (cur)
        GPRINT:Download:MAX:Download \: %2.2lf%s (max)
        GPRINT:Download:MIN:Download \: %2.2lf%s (min)
        GPRINT:Download:AVERAGE:Download \: %2.2lf%s (avg)\n
        GPRINT:Upload:LAST:Upload \: %2.2lf%s (cur)
        GPRINT:Upload:MAX:Upload \: %2.2lf%s (max)
        GPRINT:Upload:MIN:Upload \: %2.2lf%s (min)
        GPRINT:Upload:AVERAGE:Upload \: %2.2lf%s (avg)\n
 
and this is in xymonserver.cfg:
 
TEST2RRD="...,speed=ncv,..."
GRAPHS="...,speed"
NCV_speed="Download:GAUGE,Upload:GAUGE"

The test runs every 20 minutes and I can see in speed.rrd the values on top of
the file:
 
<?xml version="1.0" encoding="utf-8"?>
<!DOCTYPE rrd SYSTEM "http://oss.oetiker.ch/rrdtool/rrdtool.dtd">;
<!-- Round Robin Database Dump -->
<rrd>
        <version>0003</version>
        <step>300</step> <!-- Seconds -->
        <lastupdate>1443085347</lastupdate> <!-- 2015-09-24 11:02:27 CEST -->

        <ds>
                <name> Download </name>
                <type> GAUGE </type>
                <minimal_heartbeat>600</minimal_heartbeat>
                <min>NaN</min>
                <max>NaN</max>

                <!-- PDP Status -->
                <last_ds>13400801</last_ds>
                <value>NaN</value>
                <unknown_sec> 147 </unknown_sec>
        </ds>

        <ds>
                <name> Upload </name>
                <type> GAUGE </type>
                <minimal_heartbeat>600</minimal_heartbeat>
                <min>NaN</min>
                <max>NaN</max>

                <!-- PDP Status -->
                <last_ds>1027604</last_ds>
                <value>NaN</value>
                <unknown_sec> 147 </unknown_sec>
        </ds>

But these values are not passed down to the "archive" section of the file.
So I might be missing something or maybe my graph definition is wrong.
Can someone with a little bit more knowledge of RRD shed some light on this.
 
Thanks
Torsten

On 24/09/15 11:21, user-c862b499d9fa@xymon.invalid wrote:

Increase the heartbeat from 600 [s] to f.i. 2500 [s]. Then it is allowed 
that the script misses one pass without a hole in the graph.

Regards,
   Wim Nelis.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Wim,

thanks for the answer but how and where do I increase the heartbeat.
Do I have to create a new section in rrddefinitions.cfg?

Thanks
Torsten

On 24.09.2015 12:05, W.J.M. Nelis wrote:

Increase the heartbeat from 600 [s] to f.i. 2500 [s]. Then it is allow
ed

- -- 
+---------------------------------------------------------+

| E-mail  : user-c862b499d9fa@xymon.invalid			  |
|							  |
| Homepage: http://www.richter-it.net/			  |
+---------------------------------------------------------+
Download my public key from:
http://gpg-keyserver.de/pks/lookup?search=0x899093AC&op=get
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (MingW32)

iEYEARECAAYFAlYEGi4ACgkQ7DlmxomQk6zG2ACglNegTqec8G07goPIOUJ7F+Re
HZ0AoLeJma/arivewRWVSzD9/a2E8zr1
=e8nd
-----END PGP SIGNATURE-----

On 25 September 2015 at 01:43, Torsten Richter <user-c862b499d9fa@xymon.invalid> wrote:

Yes and no.  New RRD files are created with the parameters in
rrddefinitions.cfg.  But existing files will either need to be deleted and
re-created, or modifed using rrdtune.  See a thread on this:

http://lists.xymon.com/pipermail/xymon/2012-December/036429.html

Cheers
Jeremy

I was hoping you can give me some advice in regard to graphing data.

I have data coming in that is posted in xymon with HTML tags.  I need to parse the data out and post it into rrdtool then display it on the status page.  I am fairly new to xymon, in BB I used bb-larrd.pl to parse the data and post it into rrd.  What is the best way of doing it in xymon?

I think there are several ways of accomplishing this in xymon if I am not mistaken:


1.      As the data comes into xymon and displays on the status message, I can parse out the data that I need and send it through the rrd channel

2.      I could also within the script that generates the data include an rrd module that will create the rrd file and post the results that I need directly to the rrdtool, but then I need to include it where so it can appear on the status page

Any suggestions will be greatly appreciated.

Here is an example of the data that I want to graph.  This is how it appears on the status page.

Orders in DF staging table by Process Flag


Org

Creation Date

Process Flag

Reqs with no Orders

[red]FDC

10/19/17

P

458

[red]MDC

10/19/17

P

3249

Daniel

As you have suggested, there are several ways to do this. Xymon has a few
built-in mechanisms to support graphing, that BB didn't have. Check out the
man page for xymongraph, or the equivalent web page <
http://xymon.sourceforge.net/xymon/help/howtograph.html>; for more
information.

The key with the built-in methods is being able to present the data values
to Xymon in a structured format for it to be able to parse (eg NCV format
status message, "trends" message). If you want to handle the parsing and
graphing yourself, you can hook into the messaging stream directly (eg the
status channel) or you could create a script that periodically fetches the
current status message and parses it for the interesting data.

The way I tend to do telemetry collection and reporting is to do minimal
work on the Xymon client and most of the work on the Xymon server. You
might choose to report the status that includes a message body with a
machine-parseable format, perhaps in HTML comments if you don't want the
raw data displayed; then you have a server-side script get the status
message from the Xymon server and create a data message, which Xymon uses
to create and update RRD file(s). Another option is to arrange for the
client to add information into the client message (eg by adding a script
into the "sections" directory) and then server-side extract that section
from client message for parsing and feeding the data into Xymon.

However, it's possible to do most of the work on the client, and generate
"data" or "trends" messages (separately to any status messages you might
want the client to send), and they get injected directly into the graphing
parts of Xymon.

Hope that helps.

J

On 20 October 2017 at 03:53, LOZOVSKY, DANIEL L <user-5085da3588ee@xymon.invalid> wrote:

[image: red]FDC

10/19/17

P

458

[image: red]MDC

10/19/17

P

3249