cacti-user Mailing List for Cacti

Jose Madrid wrote:
> Dominic,
>
> Here is an export of my graph template for ucd/net - Memory Usage.
> This comes from Version 0.8.6h.
>

Thanks Jose!

> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
>

Dominic,

Here is an export of my graph template for ucd/net - Memory Usage.
This comes from Version 0.8.6h.

On 6/30/06, Dominic Marks <do...@he...> wrote:
> Hello,
>
> I managed to damage one of the standard templates by accident. I
> deleted the first item in the list, the Memory Free stack.
>
> I don't seem to be able to recreate it they way it was, although I
> tried creating it again based on the other two stacks as examples, but
> so luck so far.
>
>
> The template in question is the ucd/net Memory Template graph template.
> For reference the error I am seeing in the apache log is:
>
> ERROR: Could not make sense out of ':cdefa#FF00FF::STACK'
>
> How can I go revert the original?
>
> Thanks,
> Dominic
>
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
>


-- 
Common sense is the best distributed commodity in the world, for every
man is convinced that he is well supplied with it.
Rene Descartes "Discourse on Method"

Hello,

I managed to damage one of the standard templates by accident. I
deleted the first item in the list, the Memory Free stack.

I don't seem to be able to recreate it they way it was, although I
tried creating it again based on the other two stacks as examples, but
so luck so far.


The template in question is the ucd/net Memory Template graph template.
For reference the error I am seeing in the apache log is:

ERROR: Could not make sense out of ':cdefa#FF00FF::STACK'

How can I go revert the original?

Thanks,
Dominic

Feature....

Next major release of Cacti, "guest" access will be disabled by default.

Tony Roman
Cacti Developer

> Hello, I try to see graphs without be logged in cacti and I can do this.
>
> Example:
> http://some.domain.net/cacti/graph_image.php?local_graph_id=n&rra_id=n
>
> I tried a simple query in google to find some valid cacti
> installation, after I try to put the graph_image.php in url, and, I
> can see many graphs over internet.
>
> This is a feature of cacti or a bug?
>
> Thanks
>
> Note: I'm running Version 0.8.6h and tha *problem* is figured.
>
> --
> David Verzolla
> FreeBSD The Power to Serve
>
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job
> easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
>


-- 
Tony

Hello, I try to see graphs without be logged in cacti and I can do this.

Example:
http://some.domain.net/cacti/graph_image.php?local_graph_id=n&rra_id=n

I tried a simple query in google to find some valid cacti
installation, after I try to put the graph_image.php in url, and, I
can see many graphs over internet.

This is a feature of cacti or a bug?

Thanks

Note: I'm running Version 0.8.6h and tha *problem* is figured.

-- 
David Verzolla
FreeBSD The Power to Serve
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Thanks Larry, guess I should have a looked a little harder! :) 


Regards

Chris Fishwick
Network Administrator
Technology Services Group

Phone:  (03) 6221 1626
Fax:    (03) 6225 2424
Mobile: (0407) 330 549  - Preferred

-----Original Message-----
From: Larry Adams [mailto:lar...@co...] 
Sent: Wednesday, 28 June 2006 11:03 AM
To: 'Support requests/questions about cacti'
Subject: Re: [cacti-user] Resetting availaility figures

Chris,

You have this option today.  Simply select the hosts to reset and then from
the drop down at the bottom right hand corner, select "Clear Statistics".

Regards,

Larry Adams
Cacti Developer
The Cacti Group

> -----Original Message-----
> From: cac...@li... [mailto:cacti-user- 
> bo...@li...] On Behalf Of Fishwick, Chris
> Sent: Tuesday, June 27, 2006 8:41 PM
> To: cac...@li...
> Subject: [cacti-user] Resetting availaility figures
> 
> All,
> 
> Just wondering if anyone knows of a way to reset the availability 
> figures on the device screen?  I want to use these for tracking our 
> vendors uptime on a monthly basis, but can't at the moment unless I 
> can reset them...
> 
> I am running v0.8.6h.
> 
> Regards
> 
> Chris
> 
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job 
> easier Download IBM WebSphere Application Server v.1.0.1 based on 
> Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=1216
> 42 _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user


Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job
easier Download IBM WebSphere Application Server v.1.0.1 based on Apache
Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
cacti-user mailing list
cac...@li...
https://lists.sourceforge.net/lists/listinfo/cacti-user

Chris,

You have this option today.  Simply select the hosts to reset and then from
the drop down at the bottom right hand corner, select "Clear Statistics".

Regards,

Larry Adams
Cacti Developer
The Cacti Group

> -----Original Message-----
> From: cac...@li... [mailto:cacti-user-
> bo...@li...] On Behalf Of Fishwick, Chris
> Sent: Tuesday, June 27, 2006 8:41 PM
> To: cac...@li...
> Subject: [cacti-user] Resetting availaility figures
> 
> All,
> 
> Just wondering if anyone knows of a way to reset the availability figures
> on
> the device screen?  I want to use these for tracking our vendors uptime on
> a
> monthly basis, but can't at the moment unless I can reset them...
> 
> I am running v0.8.6h.
> 
> Regards
> 
> Chris
> 
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job
> easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user

All,

Just wondering if anyone knows of a way to reset the availability figures on
the device screen?  I want to use these for tracking our vendors uptime on a
monthly basis, but can't at the moment unless I can reset them...

I am running v0.8.6h.

Regards

Chris

Hi!

I'v a short question about setting right permissions:

/var/lib/cacti/rra ,
/var/log/cacti
is owned by cacti:users.

So good so far but it seems, that user www-data needs access these
files too if using cacti's www frontend?

Setting permissions for /var/log/cacti to
drwxr-xr-x  2 cacti users 1024 2006-06-17 21:19 /var/log/cacti/

doesn't let www frontend work properly.
Error message from cacti System Log Administration:

Warning: file(/usr/share/cacti/site/log/cacti.log): failed to open
stream: Permission denied in /usr/share/cacti/site/utilities.php on
line 115

So how should i set permissions correctly for /var/log/cacti? Which
user should do cacti's cronjob?
Do i have to consider two users for cacti (cacti for cronjob and
logfiles, www-data for logfiles)?

bye

lar...@co... wrote:

Cheers for the quick reply. Looks like it was corrupt mysql tables. 
These have been repaired and the graphs are drawing again

Thanks again for your help

Iain

> I have seen this issue before, but don't remember what the fix is.  It may be a crontab issue.  However, to isolate, you can do the following to things:
> 
> 1) First login as the polling user.
> 2) Run "php poller.php" as that user, manually and review the output.
> 3) If that does not bring you joy, then you can run the following: "php <path_cacti>/cmd.php 0 20" and review the output there as this output you will not see during normal polling.  If you are running cactid, then the method is somewhat different.
> 
> Do all of these tests with logging set to MEDIUM.  If you set it too high, you will not be able to see the tree's from the forest.
> 
> Regards,
> 
> Larry Adams
> Cacti Developer
> The Cacti Group
> 
>  -------------- Original message ----------------------
> From: Iain Conochie <ia...@uk...>
> 
>>Morning people,
>>
>>	I am having a rather strange issue with cacti. I had to reboot the 
>>server that cacti runs on for maintenance, and ever since the reboot, 
>>the graphs are no longer displaying. This seems due to the fact that the 
>>rrd files are not being updated.
>>
>>
>>	I have checked the poller.log, and i am getting many of these messages
>>
>>11:15:34 cactus01:$ tail poller.log
>>06/27/2006 10:39:58 AM - POLLER: Poller[0] Maximum runtime of 296 
>>seconds exceeded. Exiting.
>>sh: -: command not found
>>06/27/2006 10:49:58 AM - POLLER: Poller[0] Maximum runtime of 296 
>>seconds exceeded. Exiting.
>>sh: -: command not found
>>06/27/2006 10:59:58 AM - POLLER: Poller[0] Maximum runtime of 296 
>>seconds exceeded. Exiting.
>>sh: -: command not found
>>06/27/2006 11:04:59 AM - POLLER: Poller[0] Maximum runtime of 296 
>>seconds exceeded. Exiting.
>>sh: -: command not found
>>06/27/2006 11:14:58 AM - POLLER: Poller[0] Maximum runtime of 296 
>>seconds exceeded. Exiting.
>>sh: -: command not found
>>
>>
>>	Now, I saw that some of my hosts were unreachable, and i thought that 
>>this could be contributing to the timeout, so I removed these hosts and 
>>still i get these error messages.
>>
>>	Is there any kind of debug I can set on the poller so I can see where 
>>the script is getting stuck?
>>
>>Cheers
>>
>>Iain
>>
>>Using Tomcat but need to do more? Need to support web services, security?
>>Get stuff done quickly with pre-integrated technology to make your job easier
>>Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
>>http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
>>_______________________________________________
>>cacti-user mailing list
>>cac...@li...
>>https://lists.sourceforge.net/lists/listinfo/cacti-user
> 
> 
> 

Retirar da lista

Att,
=20
Francisco Ravel
fr...@tr...
Se=E7=E3o de Suporte / TRE-PA
(91) 3213-4690 / 3241-1938

Hi Larry,

Thanks for your reply: this has solved my problem!

lar...@co... wrote:
> Your problem is caused by the poller_output table getting too big for
> cacti's poller to read into memory.
> 
> There are two things you should do to fix this:
> 
> 1) Increate the maximum memory that PHP can utilize.  If you make it
> as high as a Gig (assuming you have that much in reserve), you may be
> able to get your data back.  You make this change in php.ini.  No
> reboot required, simply make the change between polls.

I did this, and restarted the poller by hand. It outputed loads of data
(because things were still in debug mode). A lot of my data has been
rescued, but unfortunately I ran into the 5 minute run-time limit of the
poller, and quite a bit of my data for the last 4 days has been lost,
alas: poller_output is now empty. And it looks like I need to run the
de-spiker on quite a few graphs.


Question:

Was this problem caused by the network outage in some way (e.g. polls
taking longer than normal) or is this something that happens to every
installation? Will the memory requirements keep growing in time, or only
as I add more hosts? Did I miss this in the installation guide?
Some 'php out of memory error' would really have helped in debugging
this, by the way.

Regards, Paul Boven.

I have seen this issue before, but don't remember what the fix is.  It may be a crontab issue.  However, to isolate, you can do the following to things:

1) First login as the polling user.
2) Run "php poller.php" as that user, manually and review the output.
3) If that does not bring you joy, then you can run the following: "php <path_cacti>/cmd.php 0 20" and review the output there as this output you will not see during normal polling.  If you are running cactid, then the method is somewhat different.

Do all of these tests with logging set to MEDIUM.  If you set it too high, you will not be able to see the tree's from the forest.

Regards,

Larry Adams
Cacti Developer
The Cacti Group

 -------------- Original message ----------------------
From: Iain Conochie <ia...@uk...>
> Morning people,
> 
> 	I am having a rather strange issue with cacti. I had to reboot the 
> server that cacti runs on for maintenance, and ever since the reboot, 
> the graphs are no longer displaying. This seems due to the fact that the 
> rrd files are not being updated.
> 
> 
> 	I have checked the poller.log, and i am getting many of these messages
> 
> 11:15:34 cactus01:$ tail poller.log
> 06/27/2006 10:39:58 AM - POLLER: Poller[0] Maximum runtime of 296 
> seconds exceeded. Exiting.
> sh: -: command not found
> 06/27/2006 10:49:58 AM - POLLER: Poller[0] Maximum runtime of 296 
> seconds exceeded. Exiting.
> sh: -: command not found
> 06/27/2006 10:59:58 AM - POLLER: Poller[0] Maximum runtime of 296 
> seconds exceeded. Exiting.
> sh: -: command not found
> 06/27/2006 11:04:59 AM - POLLER: Poller[0] Maximum runtime of 296 
> seconds exceeded. Exiting.
> sh: -: command not found
> 06/27/2006 11:14:58 AM - POLLER: Poller[0] Maximum runtime of 296 
> seconds exceeded. Exiting.
> sh: -: command not found
> 
> 
> 	Now, I saw that some of my hosts were unreachable, and i thought that 
> this could be contributing to the timeout, so I removed these hosts and 
> still i get these error messages.
> 
> 	Is there any kind of debug I can set on the poller so I can see where 
> the script is getting stuck?
> 
> Cheers
> 
> Iain
> 
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user

On 27/06/06, Paul Boven <p....@sa...> wrote:
>
> 06/27/2006 12:06:51 PM - POLLER: Poller[0] DEBUG: About to Spawn a
> Remote Process [CMD: /usr/bin/php, ARGS: -q /var/www/cacti/cmd.php 0 34]
> Waiting on 1/1 pollers.

Is SELinux enabled? It might be worth temporarily disabling it,
rebooting and seeing if that helps. Though if SELinux were the culprit
I'd have expected some pointers to this in /var/log/messages or
/var/log/secure perhaps?

Will.

Your problem is caused by the poller_output table getting too big for cacti's poller to read into memory.

There are two things you should do to fix this:

1) Increate the maximum memory that PHP can utilize.  If you make it as high as a Gig (assuming you have that much in reserve), you may be able to get your data back.  You make this change in php.ini.  No reboot required, simply make the change between polls.

2) If that does not work, then you will loose the historical data, but you can get your polling running again.  You have two options:

a) Enter into mysql console and use the following command: "truncate table poller_output;".  This way you loose historical data (since updating stopped)
b) Modify <path_cacti>/lib/poller.php to place a LIMIT statement on the selects from poller_output table.  If you set it the LIMIT to 15000, you should be able to get your data back as well.

I hope this helps.

Regards,

Larry Adams
Cacti Developer
The Cacti Group
 -------------- Original message ----------------------
From: Paul Boven <p....@sa...>
> Hi again Will, everyone,
> 
> Will McDonald wrote:
> 
> >> The graphs are still there and properly visible trough the website, they
> >> just don't get any new data added even though the poller still runs and
> >> is able to retrieve SNMP data according to the poller log.
> > 
> > Weird. Have you tried strace-ing the poller process to see if it can
> > do everything it's meant to be doing?
> 
> No, I'll give that a try - though it seems already that it never does 
> call rrdtool, but it does a lot of sql queries of course, and those 
> work. But I never see any calls to rrdtool in the cacti log from before 
> the problems.
> 
> > Have any permissions changed on the directories your RRDs live in? On
> > my Cacti system /var/www/cacti/log and /var/www/cacti/rra are both
> > owned by cacti:cacti. Assuming your poller's running out of
> > /etc/cron.d/cacti as follows...
> > 
> > */5 * * * *     cacti   php /var/www/cacti/poller.php &>/dev/null
> > 
> > It's running as cacti and would (presumably) need write permissions to
> > the logs and the RRDs.
> 
> Exactly the same here: log and rra are owned by cacti:cacti, and so are 
> the rra's. The logfile still gets updated. A file permission problem 
> would have shown up in te cacti-logs, I'd hope.
> 
> Is there any way I can increase the verbosity of a manually started 
> poller run, when the logging in the web interface is already set to 'DEBUG'?
> 
> $ php poller.php
> Content-type: text/html
> X-Powered-By: PHP/4.3.9
> 
> 06/27/2006 12:06:51 PM - POLLER: Poller[0] DEBUG: About to Spawn a 
> Remote Process [CMD: /usr/bin/php, ARGS: -q /var/www/cacti/cmd.php 0 34]
> Waiting on 1/1 pollers.
> 
> (and that's all)
> 
> Regards, Paul Boven.
> 
> 
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user

Morning people,

	I am having a rather strange issue with cacti. I had to reboot the 
server that cacti runs on for maintenance, and ever since the reboot, 
the graphs are no longer displaying. This seems due to the fact that the 
rrd files are not being updated.


	I have checked the poller.log, and i am getting many of these messages

11:15:34 cactus01:$ tail poller.log
06/27/2006 10:39:58 AM - POLLER: Poller[0] Maximum runtime of 296 
seconds exceeded. Exiting.
sh: -: command not found
06/27/2006 10:49:58 AM - POLLER: Poller[0] Maximum runtime of 296 
seconds exceeded. Exiting.
sh: -: command not found
06/27/2006 10:59:58 AM - POLLER: Poller[0] Maximum runtime of 296 
seconds exceeded. Exiting.
sh: -: command not found
06/27/2006 11:04:59 AM - POLLER: Poller[0] Maximum runtime of 296 
seconds exceeded. Exiting.
sh: -: command not found
06/27/2006 11:14:58 AM - POLLER: Poller[0] Maximum runtime of 296 
seconds exceeded. Exiting.
sh: -: command not found


	Now, I saw that some of my hosts were unreachable, and i thought that 
this could be contributing to the timeout, so I removed these hosts and 
still i get these error messages.

	Is there any kind of debug I can set on the poller so I can see where 
the script is getting stuck?

Cheers

Iain

Hi again Will, everyone,

Will McDonald wrote:

>> The graphs are still there and properly visible trough the website, they
>> just don't get any new data added even though the poller still runs and
>> is able to retrieve SNMP data according to the poller log.
> 
> Weird. Have you tried strace-ing the poller process to see if it can
> do everything it's meant to be doing?

No, I'll give that a try - though it seems already that it never does 
call rrdtool, but it does a lot of sql queries of course, and those 
work. But I never see any calls to rrdtool in the cacti log from before 
the problems.

> Have any permissions changed on the directories your RRDs live in? On
> my Cacti system /var/www/cacti/log and /var/www/cacti/rra are both
> owned by cacti:cacti. Assuming your poller's running out of
> /etc/cron.d/cacti as follows...
> 
> */5 * * * *     cacti   php /var/www/cacti/poller.php &>/dev/null
> 
> It's running as cacti and would (presumably) need write permissions to
> the logs and the RRDs.

Exactly the same here: log and rra are owned by cacti:cacti, and so are 
the rra's. The logfile still gets updated. A file permission problem 
would have shown up in te cacti-logs, I'd hope.

Is there any way I can increase the verbosity of a manually started 
poller run, when the logging in the web interface is already set to 'DEBUG'?

$ php poller.php
Content-type: text/html
X-Powered-By: PHP/4.3.9

06/27/2006 12:06:51 PM - POLLER: Poller[0] DEBUG: About to Spawn a 
Remote Process [CMD: /usr/bin/php, ARGS: -q /var/www/cacti/cmd.php 0 34]
Waiting on 1/1 pollers.

(and that's all)

Regards, Paul Boven.

On 27/06/06, Paul Boven <p....@sa...> wrote:
> Will McDonald wrote:
> > On 27/06/06, Paul Boven <p....@sa...> wrote:
> >> Just found out that last friday, the graphs on cacti stopped updating.
> >> The filestamps of the files in the rra subdirectory are all from last
> >> friday. The poller still runs, and is able to gather snmp data. I see no
> >> error messages at all in the logfile. I've tried rebuilding the
> >> poller_cache and even rebooted the machine to no avail.
> >> The system runs RHEL4, Cacti-0.8.6h-1 (built from SRPM).
> >> The graphing seems tohave stopped during a major network outage, but
> >> that was resolved the same evening - graphs still aren't updated. There
> >> is plenty of disk-space.
> >
> > Your version of RRD Tool's probably switched from 1.0.x to 1.2.x. You
> > can check with...
> >
> > $ rpm -q rrdtool
>
> rrdtool-1.0.50-3.el4.rf
>
> This has not changed recently: Install Date: Tue 02 May 2006 01:19:49 PM
> CEST

Damn. That would've been too simple I suppose. :)

> The graphs are still there and properly visible trough the website, they
> just don't get any new data added even though the poller still runs and
> is able to retrieve SNMP data according to the poller log.

Weird. Have you tried strace-ing the poller process to see if it can
do everything it's meant to be doing?

Have any permissions changed on the directories your RRDs live in? On
my Cacti system /var/www/cacti/log and /var/www/cacti/rra are both
owned by cacti:cacti. Assuming your poller's running out of
/etc/cron.d/cacti as follows...

*/5 * * * *     cacti   php /var/www/cacti/poller.php &>/dev/null

It's running as cacti and would (presumably) need write permissions to
the logs and the RRDs.

Will.

Hi Will,

Thank you for your quick reply!

Will McDonald wrote:
> On 27/06/06, Paul Boven <p....@sa...> wrote:
>> Just found out that last friday, the graphs on cacti stopped updating.
>> The filestamps of the files in the rra subdirectory are all from last
>> friday. The poller still runs, and is able to gather snmp data. I see no
>> error messages at all in the logfile. I've tried rebuilding the
>> poller_cache and even rebooted the machine to no avail.
>> The system runs RHEL4, Cacti-0.8.6h-1 (built from SRPM).
>> The graphing seems tohave stopped during a major network outage, but
>> that was resolved the same evening - graphs still aren't updated. There
>> is plenty of disk-space.
> 
> Your version of RRD Tool's probably switched from 1.0.x to 1.2.x. You
> can check with...
> 
> $ rpm -q rrdtool

rrdtool-1.0.50-3.el4.rf

This has not changed recently: Install Date: Tue 02 May 2006 01:19:49 PM 
CEST

> There are probably loads of errors in your Apache error logs,
> /var/log/httpd/error_log...

No, there are no such errors: the error log is emtpy except for 
start/stop messages.

The graphs are still there and properly visible trough the website, they 
just don't get any new data added even though the poller still runs and 
is able to retrieve SNMP data according to the poller log.

Regards, Paul Boven.

On 27/06/06, Paul Boven <p....@sa...> wrote:
> Just found out that last friday, the graphs on cacti stopped updating.
> The filestamps of the files in the rra subdirectory are all from last
> friday. The poller still runs, and is able to gather snmp data. I see no
> error messages at all in the logfile. I've tried rebuilding the
> poller_cache and even rebooted the machine to no avail.
> The system runs RHEL4, Cacti-0.8.6h-1 (built from SRPM).
> The graphing seems tohave stopped during a major network outage, but
> that was resolved the same evening - graphs still aren't updated. There
> is plenty of disk-space.

Your version of RRD Tool's probably switched from 1.0.x to 1.2.x. You
can check with...

$ rpm -q rrdtool

There are probably loads of errors in your Apache error logs,
/var/log/httpd/error_log...

ERROR: Garbage ':01:51 To 2006/06/23 12:01:51\c' after command:
COMMENT:From 2006/06/22 12:01:51 To 2006/06/23 12:01:51\c
ERROR: Garbage ':01:51 To 2006/06/23 12:01:51\c' after command:
COMMENT:From 2006/06/22 12:01:51 To 2006/06/23 12:01:51\c

Log in to Cacti and got to Configuration -> Settings and switch
RRDTool Utility Version to match that of the installed RRD Tool.

Will.

Hi everyone,


Just found out that last friday, the graphs on cacti stopped updating. 
The filestamps of the files in the rra subdirectory are all from last 
friday. The poller still runs, and is able to gather snmp data. I see no 
error messages at all in the logfile. I've tried rebuilding the 
poller_cache and even rebooted the machine to no avail.
The system runs RHEL4, Cacti-0.8.6h-1 (built from SRPM).
The graphing seems tohave stopped during a major network outage, but 
that was resolved the same evening - graphs still aren't updated. There 
is plenty of disk-space.

Any hints on where to look, what to do?

Regards, Paul Boven.

    Hi everybody,

    Sorry for perhaps such trivial question...

    I want to graph same datas for hundred hosts...I created a host 
template describing the profile of what i want to get...

    But now, am i obliged to click hundred times on the "add device" 
link and just changing the hostname....or do I have a way to batch such 
actions (mysql script? bash script?...)?

    Thanks for any help
    Regards
    Paul.

-- 
Paul TAVERNIER
Network engineer
Rectorat de ROUEN

Hollywood Intermediate Inc.
(SYM : H Y W I) 

HOT news just released

Current Sh Price : $ 0.65
 
Follow the performance of this company, it is a real gold mine
try and miss this 

HYWI has performed like clockwork every time 

CO OverView
H o l l y w o o d  I n t e r m e d i a t e provides a proprietary technology of Digital Intermediate services to feature filmmakers for post-production for film mastering and restoration. This technology gives the filmmakers total creative control over the look of their productions. Whether shooting on film or acquiring in HD or SD video, H o l l y w o o d  I n t e r m e d i a t e puts a powerful cluster of digital tools at the director's disposal to achieve stunning results on the big screen. Matchframe Digital Intermediate, a division of H o l l y w o o d  I n t e r m e d i a t e, Inc., packages a full array of post-production services with negative handling expertise and cost-effective 2K digital intermediate and 35mm film out systems. The Digital Intermediate process eliminates current post-production redundancies by creating a single high-resolution master file from which all versions can be made, including all theatrical and High Definition formats. By creating a single master file with resolution higher than the current High Definition broadcast standards, the DI master file enables cinema and television distributors to extract and archive all current and future cinema and television formats including Digital Cinema, Television and High Definition.


Red H0t News:

Hollywood Intermediate Announces Completion of Work on Several Digital Intermediates by Senior Colorist Julius Friede GLENDALE, CA--(MARKET WIRE)--Jun 21, 2006 -- Hollywood Intermediate, Inc. (Other OTC:HYWI.PK - News) a provider of digital intermediate film mastering services announced today that Julius Friede, senior colorist at its Matchframe Digital Intermediate division, has completed work on several digital intermediates in the last three months, including "Journey to the End of the Night," directed by Eric Eason, "The Sensation of Sight," directed by Aaron Wiederspahn and shot by Christophe Lanzenberg, and "Beautiful Ohio," directed by Chad Lowe, and shot by Steve Kazmierski.

H o l l y w o o d  I n t e r m e d i a t e a provider of digital intermediate film mastering services, announced today that that its Matchframe Digital Intermediate (MDI) division is completing a digital intermediate for Chad Lowe's directorial debut, "Beautiful Ohio," starring William Hurt and Rita Wilson. READ MORE THIS IS HUGE

H o l l y w o o d  I n t e r m e d i a t e Expands the Creative Palette for Independent Filmmakers GLENDALE, CA--(MARKET WIRE)--May 31, 2006 -- H o l l y w o o d  I n t e r m e d i a t e, Inc. A provider of digital intermediate film mastering services, announced today that its Matchframe Digital Intermediate division is currently providing full digital intermediate services for Super 16MM productions.

H o l l y w o o d  I n t e r m e d i a t e, Inc. (H_Y_W_I.PK - News), a provider of digital intermediate film mastering services, announced that High Definition preview masters as part of its normal digital intermediate service offerings and workflow.

"Typically, in current post-production workflow, HD dailies masters are edited into high quality preview masters including color timing, dirt removal, opticals and visual effects," said David Waters, H o l l y w o o d  I n t e r m e d i a t e president. "Unfortunately, none of these processes translate to the theatrical release of the film as they must all be duplicated or repeated in either a higher resolution digital format, or photo chemical process."

H o l l y w o o d  I n t e r m e d i a t e gives Motion Picture producers the ability to scan their selected original camera negative at 2k or 4k film resolution, conform a high resolution digital master for theatrical and broadcast release including dirt removal, opticals and visual effects, and output a High Definition preview master to be used for preview screenings and focus groups that can be deployed in any worldwide theater location.

"The challenge for completing the final editorial decisions on a motion picture are balanced between the ability to display the highest resolution picture for a test audience, and the costs and time in having to re-master your film based on a test audience response," said Jim Delany, H o l l y w o o d  I n t e r m e d i a t e COO.

DO your Due Diligence and you'll see what we are talking about when it comes to H-Y-W-I.PK

-----------------------
Sweet as apple pie.   Put to bed with a shovel. Red as a beet.  Say it with flowers.   Thick as a brick.   Put that in your pipe and smoke it.   A tree does not move unless there is wind.   Watch and wait.  Water doesn't run uphill.   She's the apple of my eye.   Up one side and down the other.   A rose by any other name would smell as sweet. Read the tea leaves. That's a whole new can of worms.   You have to separate the chaff from the wheat. Want my place in the sun.  Spring forward fall back. The season of goodwill.   Shall I compare thee to a summer's day. She's a mother hen.   She's the apple of my eye.   Plant kindness and gather love.   Water it down.  Spring to mind.  

Sweet as honey.   Two peas in a pod.  The sharper is the berry, the sweeter is the wine. Watch and wait.  Shit happens.   Timber! Stop, look and listen. A place in the sun.   We'll cross that bridge when we come to it.  That's a whole new can of worms.   Where man is not nature is barren. Stop and smell the roses. When pigs fly.  Walking on cloud nine.

Hi!

cacti poller is not running.
Starting it manually gives error message:

# /usr/share/cacti/site/poller.php

Warning: chdir(): No such file or directory (errno 2) in
/usr/share/cacti/site/poller.php on line 90


# strace /usr/share/cacti/site/poller.php

read(3, "\vpoller_item\7rrd_num\3\2\0\0\1\1\3!\0\0", 30) = 30
read(3, "\1\0\0\10", 4)                 = 4
read(3, "\376", 1)                      = 1
read(3, "\1\0\0\t", 4)                  = 4
read(3, "\376", 1)                      = 1
time(NULL)                              = 1151270949
rt_sigprocmask(SIG_BLOCK, [CHLD], [], 8) = 0
rt_sigaction(SIGCHLD, NULL, {SIG_DFL}, 8) = 0
rt_sigprocmask(SIG_SETMASK, [], NULL, 8) = 0
nanosleep({1, 0},

[LOOP]

What could be wrong here?
As result i don't get any graphs in www frontend.

bye