headers
Tomas | 8 Oct 2004 13:58
Picon

Compiling rrd tools to custom application

Hello,
I have compiled in rrd tool into my application because calling of
rrrtool update executable takes a lot of time. I found following
problems:

1) there are no exported symbols in object archive library librrd.a from
rrdtool-devel-1.0.41-1.8.0.i386.rpm. Hence is useless and it's necesary
recreate it from compiled rrdtool.

2) there is no API function for database update except rrd_update
(rrd_update.c). But this function is not handy because it is necessary
to convert values to strings and rrd_update will convert back to native
types. In addition template resolving causes a critical error. I think
that is worth writing new API function rrd_api_update that takes native
parametres and rewrite rrd_update to decode string (argv) and pass it to
rrd_api_update. There is another problem if custom function uses own
memory allocation routines and rrd library standard ones.

3) Alignment critical structures should be safed to be portable across
executables. One custom software should update database and using native
rrdtool can create graphs, prepare databases etc. For example my
software used "-m align-doubles" and it causes incompatability of
structures read from rrd database (in my case in stat_head_t). I think
some #pragma align for critical structures should prevent such problems.

Tomas Mandys              
http://www.2p.cz

--
Unsubscribe mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=unsubscribe
(Continue reading)

Permalink | Reply | 0 comments |
headers
Sunil Modi (IT | 8 Oct 2004 16:37

RRD eating values

I generated some backlogged data from January of 2004 into an RRD.  I created the RRD one month at a time
(tedious but RRD had problems taking all the data in one large update file).  I had run a test case of the first
month to see how the data would look and it looked just fine when compared to excel graphs.  
When I finished making the RRD and putting in all 9 months of data, I generated some graphs, and saw that the
graphs are different and the peak points are different.  The single Month Graph had a peak of 100, and the 9
Month graph had a peak of 60.  What happened to the peak of 100?  Should I make separate RRD's for each month to
be more accurate?

Thanks,

Sunil Modi
IT-Server Administration
(734) 477-1570
smodi <at> bordersgroupinc.com

--
Unsubscribe mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
Permalink | Reply | 1 comment |
headers
David M. Grimes | 8 Oct 2004 17:18
Favicon

Re: RRD eating values

On Fri, Oct 08, 2004 at 10:37:55AM -0400, Sunil Modi (IT) wrote:
> I generated some backlogged data from January of 2004 into an RRD.  I created the RRD one month at a time
(tedious but RRD had problems taking all the data in one large update file).  I had run a test case of the first
month to see how the data would look and it looked just fine when compared to excel graphs.  
> When I finished making the RRD and putting in all 9 months of data, I generated some graphs, and saw that the
graphs are different and the peak points are different.  The single Month Graph had a peak of 100, and the 9
Month graph had a peak of 60.  What happened to the peak of 100?  Should I make separate RRD's for each month to
be more accurate?

The effect you're seeing is the result of having too much data density to
fit on the graph you are generating.  For a 9 month graph at default
resolution (400x100) you've got about 16 hours worth of data that get
"rolled up" to form a pixel on the graph.  If you're using the AVERAGE CF
for the archive, you'll be averageing 16 hours worth of data, which is
naturally going to remove the "peak".

If you're using the CVS snapshot (development branch) after about
09-27-2004, you can use this new option on the DEF: when you draw the graph:

DEF:x=rrd:ds:step=<step override>:reduce=<CF>

And the step override should be the granularity of your "highest density"
RRA, and the CF for reduce would be MAX to achieve what you're trying to
see.

Hope that helps,

 --Dave Grimes

>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Tobias Oetiker | 8 Oct 2004 23:14
Picon
Picon
Favicon

Re: Patch to force displaying [HV]RULE legend

Sep 20 Antoine Delvaux wrote:

> Hello,
> Long time delayed, but here is now the corresponding patch
> (--force-rules-legend) for the development branch.  This patch is
> against version 2004-09-20.
>
> I hope it can be included from now on.  Thanks.

your patch is in ... sorry for the delay ... I was offline

tobi
>
> Antoine.
>
>

--

-- 
 ______    __   _
/_  __/_  / /  (_) Oetiker  <at>  ISG.EE, ETL F24.2, ETH, CH-8092 Zurich
 / // _ \/ _ \/ /  System Manager, Time Lord, Coder, Designer, Coach
/_/ \.__/_.__/_/   http://people.ee.ethz.ch/oetiker +41(0)44-632-5286

--
Unsubscribe mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
(Continue reading)

Permalink | Reply | 1 comment |
headers
Ricardo Jorge | 9 Oct 2004 02:37
Picon

Patch to fix core dump on Tcl module

Dear developers !

    I am sending the patch that fix core dump when running rrdtool from Tcl.

    I hope that help others. Thank you.

             Ricardo Jorge

================== diff.txt ================

96c96
<     getopt_cleanup(argv, argv2);
---
 >     getopt_cleanup(argc, argv2);
124c124
<     getopt_cleanup(argv, argv2);
---
 >     getopt_cleanup(argc, argv2);
152c152
<     getopt_cleanup(argv, argv2);
---
 >     getopt_cleanup(argc, argv2);
197c197
<     getopt_cleanup(argv, argv2);
---
 >     getopt_cleanup(argc, argv2);
242c242
<     getopt_cleanup(argv, argv2);
---
 >     getopt_cleanup(argc, argv2);
(Continue reading)

Permalink | Reply | 0 comments |
headers
Didier | 10 Oct 2004 18:03
Picon
Favicon

Ask for rrdtool.exe to test new TREND CDEF function

Ask for rrdtool.exe to test new TREND CDEF function

Hello,

 Can anybody send me CVS release already compiled of  the rrdtool.exe
forWindows (XP)
 I'd like to test the new TREND CDEF fucntion that   David M. Grimes has
integrated.
 It has been integrated from rrdtool-cvs-snapshot (2004-09-27).

     Thanks a lot

Didier

--
Unsubscribe mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
Permalink | Reply | 0 comments |
headers
Christopher Snell | 20 Oct 2004 22:03
Picon

gdImageCopyResampled

Hi,

I was browsing through the libgd documentation and saw a (relatively)
new function called gdImageCopyResampled.  This takes a gdImagPtr and
smoothly (or so it says) resamples it to a smaller size.  This very
useful functionality.  Sometimes its nice to have resampled
graphs--you can display more graph in less area than you could by
simply shrinking the graph size within RRDtool.  Currently, I'm doing
this via calls via ImageMagick's library but this is quite slow
because we are going from GD -> PNG -> I.M. Format -> PNG.  I'm
working on a patch to add  --resample-width and --resample-height to
RRDtool but I'm a very slow and not-very-good C coder.  Perhaps
someone could beat me to the punch?

Chris

gdImageCopyResampled docs:

http://www.boutell.com/gd/manual2.0.28.html#gdImageCopyResampled

Example resampled image (done on-the-fly with ImageMagick).

http://chrissnell.com:48000/graphs/?graph_type=temperaturec&start=8000&small=1

--
Unsubscribe mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
(Continue reading)

Permalink | Reply | 0 comments |
headers
Haroon Rafique | 21 Oct 2004 19:43
Picon
Picon
Favicon

inconsistency in src/parsetime.c rrdtool 1.0.49 (fwd)

I didn't get any bites on the rrd-users list, so trying the developers 
list.

--
Haroon Rafique
<haroon.rafique <at> utoronto.ca>

---------- Forwarded message ----------
Date: Fri, 1 Oct 2004 10:49:46 -0400 (EDT)
From: Haroon Rafique <haroon.rafique <at> utoronto.ca>
To: rrd-users <at> list.ee.ethz.ch
Subject: [rrd-users] inconsistency in src/parsetime.c rrdtool 1.0.49

Hi,

I noticed an inconsistency in the documentation in src/parsetime.c. I
looked in the contrib directory and there is a trytime directory which
helped me isolate the anomaly a little further.

According to the at style time specification notes (from man rrdfetch),
-6m should be interpreted as -6 months. However, that's not the case
currently. It is interpreted as -6 minutes.

Try:

./contrib/trytime/trytime -s -6m -e -5m
The time now is: Fri Oct  1 10:35:12 2004

Start time was specified as: '-6m',
for me this means: Fri Oct  1 10:29:12 2004 EDT (or 1096640952 sec since
(Continue reading)

Permalink | Reply | 0 comments |
headers
Haroon Rafique | 21 Oct 2004 19:44
Picon
Picon
Favicon

rrdtool 1.0.49 return error code (fwd)

Second message that I didn't get any replies on the rrd-users list, so 
trying the developers list.

--
Haroon Rafique
<haroon.rafique <at> utoronto.ca>

---------- Forwarded message ----------
Date: Fri, 1 Oct 2004 11:15:48 -0400 (EDT)
From: Haroon Rafique <haroon.rafique <at> utoronto.ca>
To: rrd-users <at> list.ee.ethz.ch
Subject: [rrd-users] rrdtool 1.0.49 return error code

Hi listers,

I have a question regarding return codes from rrdtool 1.0.49.

If I execute:

  	rrdtool update /blah/nonexistent/file/path

Shouldn't the value of $? afterwards be non-zero.

Here's what I get when executing:

prompt$	rrdtool update /blah/nonexistent/file/path; echo $?
ERROR: Not enough arguments
0

Basically, I'm implying that the error handling from the rrd_*.c files
(Continue reading)

Permalink | Reply | 0 comments |
headers
Michael Haase | 28 Oct 2004 19:09
Picon

difference rrd files Linux/Windows

Hi,

how are the rrd files different in Linux and Windows version?
I have the problem that I want to read rrd files from a Linux server and
work with them on a windows client. And I'd prefer not to install something
on the server to convert the files there by request to xml, but I want to
convert them on the Windows client (for an own Java program). For that I
need to know the difference. Is there any possibility?

Thanks,
Michael

--
Unsubscribe mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:rrd-developers-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/rrd-developers
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
Permalink | Reply | 0 comments |

Gmane