Result from CMD not valid. Partial Result:

[GoogleCodeExporter]

What steps will reproduce the problem?
1. Install, setup, run...
2. Receive errors in log file

What is the expected output? What do you see instead?
Erm... I expect graphs with values on them? Instead, I see graphs with lots
of "nan" values, and lots of errors in the log:

09/29/2008 09:12:01 PM - CMDPHP: Poller[0] Host[1] DS[51] WARNING: Result
from CMD not valid. Partial Result: 

What version of the product are you using? On what operating system?
I've tried the mysql-cacti-templates 1.0.0 .tar.gz, as well as the r60 from
SVN. Other software: Fedora 9 x64, MySQL 5.0.67-1, PHP 5.2.6-2, Cacti 0.8.7b

Please provide any additional information below.
M-C-T is setup to use the same MySQL user that Cacti uses (not root); have
granted SUPER and PROCESS privileges.

I tried to make a new template for 0.8.7b, but I got this from the Perl script:
No such version '0.8.7b', try one of 0.8.4 0.8.5 0.8.5a 0.8.6 0.8.6a 0.8.6b
0.8.6c 0.8.6d 0.8.6e 0.8.6f 0.8.6g 0.8.6h 0.8.6i 0.8.6j at make-template.pl
line 547.

If I was getting no errors in the log, then I would assume the problem lies
elsewhere, but I'm thinking it's something to do with the versions...

I get no notices/warnings/errors when I manually run the PHP script from
the command line.

Original issue reported on code.google.com by f...@t-y.co.uk on 29 Sep 2008 at 8:36

[GoogleCodeExporter]

I was getting the same "Partial Result" message too.  The issue is that the 
output 
is longer than Cacti's willing to accept.  The reason that it works from some 
people 
and not others is that over time the numbers getting higher and use more 
character 
placements.  This would work if the output parameter names were shortened to 2 
characters.

Original comment by danflore...@gmail.com on 7 Oct 2008 at 6:50

[GoogleCodeExporter]

That's possible, but in the SVN version all the output parameters appear to 
have been
shortened to 2 or 3 characters already... and I still get the same error on 
that.

Original comment by f...@t-y.co.uk on 7 Oct 2008 at 7:46

[GoogleCodeExporter]

That's great.  I was still referencing the code available via the download 
page.  
Thanks.

Original comment by danflore...@gmail.com on 7 Oct 2008 at 8:09

[GoogleCodeExporter]

I'm now getting the same graphing inconsistency as freddyty.  My setup is 
CentOS 4.4 
with Cacti 0.8.6j.  The script itself returns information consistently.  Not 
sure 
where the breakdown is, but the Cacti log reports "WARNING: Result from SCRIPT 
not 
valid. Partial Result: a0:" ...

Original comment by danflore...@gmail.com on 9 Oct 2008 at 6:58

[GoogleCodeExporter]

Run the script from the command-line as follows and report what it says:

<script....> | wc -c

Original comment by baron.schwartz on 9 Oct 2008 at 9:53

[GoogleCodeExporter]

I think the fixes for issue 28 will address this problem.

It is not backwards compatible, though.  You'll have to upgrade your templates, 
not
just the PHP script.  I disclaim any responsibility for that -- this was forced 
by
yet another Cacti data model oddity.

Original comment by baron.schwartz on 14 Oct 2008 at 7:13

• Changed state: Fixed

[GoogleCodeExporter]

I've also noticed that when I run the cmd.php poller from the command line, I 
get
alot of:

[me@box cacti]$ sudo -u nagios php -q cmd.php 
sh: path_php_binary: command not found
07/29/2009 11:22:34 PM - CMDPHP: Poller[0] Host[52] DS[665] WARNING: Result 
from CMD
not valid.  Partial Result: U
sh: path_php_binary: command not found
07/29/2009 11:22:34 PM - CMDPHP: Poller[0] Host[52] DS[664] WARNING: Result 
from CMD
not valid.  Partial Result: U
sh: path_php_binary: command not found

I'm running cacti-0.8.7d-1.fc9.noarch.

Original comment by erik....@gmail.com on 30 Jul 2009 at 4:26

[GoogleCodeExporter]

  Figured out there was some issue where the <>'s signs were being omitted from the
data_input tables. I added the necessary brackets, stopped getting errors in the
cacti log, but still cant seem to get anything graphed even though its 
returning data.

Original comment by erik....@gmail.com on 30 Jul 2009 at 5:50