cacti-user Mailing List for Cacti (Page 13)

Did you rebuild the poller cache?
System Utilities -> Rebuild poller cache.

Paul

On 24-12-14 14:06, Ben .T.George wrote:
> I have installed cacti on redhat machine and i configures spine as poller.
> 
> by default localhost was there on devices list. if i click on localhost,
> the graph image is there but no plots on that.
> 
> as i am very new to this, i don't know how to troubleshot this
> 
> please hlep me to fix this issue.
> 
> Thanks & Regards,
> Ben
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming! The Go Parallel Website,
> sponsored by Intel and developed in partnership with Slashdot Media, is your
> hub for all things parallel software development, from weekly thought
> leadership blogs to news, videos, case studies, tutorials and more. Take a
> look and join the conversation now. http://goparallel.sourceforge.net
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
> 

I have installed cacti on redhat machine and i configures spine as poller.

by default localhost was there on devices list. if i click on localhost,
the graph image is there but no plots on that.

as i am very new to this, i don't know how to troubleshot this

please hlep me to fix this issue.

Thanks & Regards,
Ben

Ignacio,

What about creating a new data template for each OID you want to get?  You
can then reference those data templates in your graphs and graph templates.

You could create a new data input method for each OID, but there's already
one present for SNMP so I would just use that one.

I don't have much experience yet with data queries, but I believe they are
used when you want to create multiple graphs of the same type based on some
external data.  For example, the "SNMP - Get Mounted Partitions" data query
is associated with the "Host MIB - Available Disk Space" graph template.
When you add that graph template to a host and then create graphs you'll
see that Cacti adds an available disk graph for each of the partitions the
data query found.


On Thu, Dec 18, 2014 at 11:15 AM, Ignacio Verona <iv...@es...
> wrote:

> Hello all,
>
> I looks like indexes are a quite common question but after googling for a
> long time I've not been able to find a solution. I've a set of oids:
>
> $ snmpwalk -v2c -c vCPESecret 192.168.27.112 1.3.6.1.4.1.119.2.3.237.1 -On
> .1.3.6.1.4.1.119.2.3.237.1.1.1.1.0 = INTEGER: operating(3)
> .1.3.6.1.4.1.119.2.3.237.1.1.3.1.1.2.1 = INTEGER: 0
> .1.3.6.1.4.1.119.2.3.237.1.1.3.1.1.3.1 = INTEGER: 40000
>
> from a proprietary MIB. None of them is an index and I don't know how to
> query those 3 values. Should I use a Data Query? Create a new Data Input?
> Multiple data queries, one for each oid?
>
> any help will be really appreciated as I've been trying everything I saw
> for the whole day with no luck.
>
> Thanks,
> Ignacio.
>
> ------------------------------------------------------------------------------
> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> with Interactivity, Sharing, Native Excel Exports, App Integration & more
> Get technology previously reserved for billion-dollar corporations, FREE
>
> http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
>



-- 


*Damian Wiest**DevOps Engineer*

Apervita by Pervasive Health
One North LaSalle Street, Suite 1825
Chicago, IL 60602 USA
+1 312 257 2967

<http://www.apervita.net/>

<http://www.apervita.com/>  <https://twitter.com/apervita>
<http://www.linkedin.com/company/pervasive-health>

-- 


Confidentiality Notice: This e-mail transmission may contain confidential 
or legally privileged information that is intended only for the individual 
or entity named in the e-mail address. If you have received this 
communication in error, please notify me by return e-mail, and destroy this 
communication and all copies thereof, including any attachments. 

Apervita ® is a registered trademark of Pervasive Health Inc.

 

Hello all,

I looks like indexes are a quite common question but after googling for a
long time I've not been able to find a solution. I've a set of oids:

$ snmpwalk -v2c -c vCPESecret 192.168.27.112 1.3.6.1.4.1.119.2.3.237.1 -On
.1.3.6.1.4.1.119.2.3.237.1.1.1.1.0 = INTEGER: operating(3)
.1.3.6.1.4.1.119.2.3.237.1.1.3.1.1.2.1 = INTEGER: 0
.1.3.6.1.4.1.119.2.3.237.1.1.3.1.1.3.1 = INTEGER: 40000

from a proprietary MIB. None of them is an index and I don't know how to
query those 3 values. Should I use a Data Query? Create a new Data Input?
Multiple data queries, one for each oid?

any help will be really appreciated as I've been trying everything I saw
for the whole day with no luck.

Thanks,
Ignacio.

Hi!

I have a problem with SNMP queries and HP 10500 systems with different 
slots. I have found an XML script that I could customize.

The OID index query gives the following result:
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.0 = INTEGER: 0
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.1 = INTEGER: 1
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.4 = INTEGER: 4
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.5 = INTEGER: 5
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.9 = INTEGER: 9
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.10 = INTEGER: 10
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.11 = INTEGER: 11
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.12 = INTEGER: 12
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotIndex.1.13 = INTEGER: 13

With this starting point I can query the CPU usage or memory. My CPU 
usage walk looks like:
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.0 = INTEGER: 5
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.1 = INTEGER: 2
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.4 = INTEGER: 0
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.5 = INTEGER: 0
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.9 = INTEGER: 1
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.10 = INTEGER: 3
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.11 = INTEGER: 3
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.12 = INTEGER: 3
HH3C-LSW-DEV-ADM-MIB::hh3cLswSlotCpuRatio.1.13 = INTEGER: 3

But I have problems with the temperature. The result for this query looks 
like:
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.0.temperature = INTEGER: 49
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.1.temperature = INTEGER: 29
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.4.temperature = INTEGER: 23
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.5.temperature = INTEGER: 24
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.9.temperature = INTEGER: 33
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.10.temperature = INTEGER: 30
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.11.temperature = INTEGER: 30
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.12.temperature = INTEGER: 30
HH3C-LswDEVM-MIB::hh3cDevMSlotEnvironmentValue.1.13.temperature = INTEGER: 29

Here the slot number is not the last number in the OID. So I don’t get 
any results in cacti (always 0) as opposed to the CPU or memory queries.

Has anyone an idea how do I solve this problem?

Shade and sweet water!

	Stephan

-- 
| Stephan Seitz          E-Mail: st...@fs... |
| Public Keys: http://fsing.rootsland.net/~stse/keys.html |

Thank you, Reinhard. 

I'll try to fix the script myself. I noticed that, this function is used
by many plugintemplates; will try to see the different.

Yes, it's true, the thread where i got the template from is quite (back
to 2006). 

May I further ask if anyone on the list knows of any better template for
this? 
I would like to graphmonitor traffic of:
1. BGP peers by their AS numbers 
2. Hosts in the LAN by their MAC addresses

Thanks again, Reinhard; I've always read this list and see you always
are very active and helpful.

Regards,
Khem

On សុ, 2014-12-05 at 08:56 +0100, Gandalf wrote:
> This seems to be an outdated resource. It has been quite some time, that the
> snmpwalk interface for Cacti has been extended for full SNMP V3 capabilities,
> resulting in some new parameters. Studying the new parameters, it is likely a
> tiny task to change the script you are talking about.
> Reinhard
> 
> Am 05.12.2014 um 05:07 schrieb Khemara Lyn:
> > Dear All,
> > 
> > I have downloaded the template "traffic between 2 peers" from this
> > thread:
> > 
> > http:forums.cacti.netviewtopic.php?f=12&t=13931&sid=23c5addd956c18bc258141568462be82&start=15
> > 
> > and installed on my Cacti-0.8.8b engine.
> > 
> > It seems to work; however when I run the script:
> > 
> > php .ss_host_mac.php x.x.x.x 0 "2:161:10:comm_string:::[MD5]:None"
> > query index
> > 
> > It gives me the with the following warnings:
> > 
> > "PHP Warning:  Missing argument 10 for cacti_snmp_walk(), called
> > in varlibcactiscriptsss_host_mac.php on line 98 and defined
> > in usrsharecactilibsnmp.php on line 257
> > 
> > Warning: Missing argument 10 for cacti_snmp_walk(), called
> > in varlibcactiscriptsss_host_mac.php on line 98 and defined
> > in usrsharecactilibsnmp.php on line 257
> > PHP Notice:  Undefined variable: context
> > in usrsharecactilibsnmp.php on line 292
> > 
> > Notice: Undefined variable: context in usrsharecactilibsnmp.php on
> > line 292"
> > 
> > May I ask: how can i correct the script to get rid of the those
> > warnings. 
> > 
> >>From a bit of search on Google, I knew that, it is due to different
> > version. The template I use was for old version of Cacti. 
> > How can I adjust it to comply with the new version?
> > 
> > Hope for your kind reply and help, anyone!
> > 
> > Or better yet, is there a template to monitor traffic of BGP peers by
> > their AS numbers or by MAC address? I'm looking for something like that.
> > Sorry, this should placed in a different thread; but since it is a bit
> > related, i'd like to ask it here also.
> > 
> > Regards,
> > Khem
> > 
> > 
> > 
> > 
> > 
> > ------------------------------------------------------------------------------
> > Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> > from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> > with Interactivity, Sharing, Native Excel Exports, App Integration & more
> > Get technology previously reserved for billion-dollar corporations, FREE
> > http:pubads.g.doubleclick.netgampadclk?id=164703151&iu=4140ostg.clktrk
> > 
> > 
> > 
> > _______________________________________________
> > cacti-user mailing list
> > cac...@li...
> > https:lists.sourceforge.netlistslistinfocacti-user
> > 
> 
> ------------------------------------------------------------------------------
> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> with Interactivity, Sharing, Native Excel Exports, App Integration & more
> Get technology previously reserved for billion-dollar corporations, FREE
> http:pubads.g.doubleclick.netgampadclk?id=164703151&iu=4140ostg.clktrk
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https:lists.sourceforge.netlistslistinfocacti-user

This seems to be an outdated resource. It has been quite some time, that the
snmpwalk interface for Cacti has been extended for full SNMP V3 capabilities,
resulting in some new parameters. Studying the new parameters, it is likely a
tiny task to change the script you are talking about.
Reinhard

Am 05.12.2014 um 05:07 schrieb Khemara Lyn:
> Dear All,
> 
> I have downloaded the template "traffic between 2 peers" from this
> thread:
> 
> http://forums.cacti.net/viewtopic.php?f=12&t=13931&sid=23c5addd956c18bc258141568462be82&start=15
> 
> and installed on my Cacti-0.8.8b engine.
> 
> It seems to work; however when I run the script:
> 
> php ./ss_host_mac.php x.x.x.x 0 "2:161:10:comm_string:::[MD5]:None"
> query index
> 
> It gives me the with the following warnings:
> 
> "PHP Warning:  Missing argument 10 for cacti_snmp_walk(), called
> in /var/lib/cacti/scripts/ss_host_mac.php on line 98 and defined
> in /usr/share/cacti/lib/snmp.php on line 257
> 
> Warning: Missing argument 10 for cacti_snmp_walk(), called
> in /var/lib/cacti/scripts/ss_host_mac.php on line 98 and defined
> in /usr/share/cacti/lib/snmp.php on line 257
> PHP Notice:  Undefined variable: context
> in /usr/share/cacti/lib/snmp.php on line 292
> 
> Notice: Undefined variable: context in /usr/share/cacti/lib/snmp.php on
> line 292"
> 
> May I ask: how can i correct the script to get rid of the those
> warnings. 
> 
>>From a bit of search on Google, I knew that, it is due to different
> version. The template I use was for old version of Cacti. 
> How can I adjust it to comply with the new version?
> 
> Hope for your kind reply and help, anyone!
> 
> Or better yet, is there a template to monitor traffic of BGP peers by
> their AS numbers or by MAC address? I'm looking for something like that.
> Sorry, this should placed in a different thread; but since it is a bit
> related, i'd like to ask it here also.
> 
> Regards,
> Khem
> 
> 
> 
> 
> 
> ------------------------------------------------------------------------------
> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> with Interactivity, Sharing, Native Excel Exports, App Integration & more
> Get technology previously reserved for billion-dollar corporations, FREE
> http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
> 
> 
> 
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
> 

Dear All,

I have downloaded the template "traffic between 2 peers" from this
thread:

http:forums.cacti.netviewtopic.php?f=12&t=13931&sid=23c5addd956c18bc258141568462be82&start=15

and installed on my Cacti-0.8.8b engine.

It seems to work; however when I run the script:

php .ss_host_mac.php x.x.x.x 0 "2:161:10:comm_string:::[MD5]:None"
query index

It gives me the with the following warnings:

"PHP Warning:  Missing argument 10 for cacti_snmp_walk(), called
in varlibcactiscriptsss_host_mac.php on line 98 and defined
in usrsharecactilibsnmp.php on line 257

Warning: Missing argument 10 for cacti_snmp_walk(), called
in varlibcactiscriptsss_host_mac.php on line 98 and defined
in usrsharecactilibsnmp.php on line 257
PHP Notice:  Undefined variable: context
in usrsharecactilibsnmp.php on line 292

Notice: Undefined variable: context in usrsharecactilibsnmp.php on
line 292"

May I ask: how can i correct the script to get rid of the those
warnings. 

>From a bit of search on Google, I knew that, it is due to different
version. The template I use was for old version of Cacti. 
How can I adjust it to comply with the new version?

Hope for your kind reply and help, anyone!

Or better yet, is there a template to monitor traffic of BGP peers by
their AS numbers or by MAC address? I'm looking for something like that.
Sorry, this should placed in a different thread; but since it is a bit
related, i'd like to ask it here also.

Regards,
Khem

Release of Cacti 0.8.8c

We the Cacti Group are proud to release the following:

    Cacti 0.8.8c
    Spine 0.8.8c

Important Security Fixes

 * CVE-2013-5588 - XSS issue via installer or device editing
 * CVE-2013-5589 - SQL injection vulnerability in device editing
 * CVE-2014-2326 - XSS issue via CDEF editing
 * CVE-2014-2327 - Cross-site request forgery (CSRF) vulnerability
 * CVE-2014-2328 - Remote Command Execution Vulnerability in graph export
 * CVE-2014-4002 - XSS issues in multiple files
 * CVE-2014-5025 - XSS issue via data source editing
 * CVE-2014-5026 - XSS issues in multiple files

Important Updates

 * New graph tree view
 * Updated graph list and graph preview
 * Refactor graph tree view to remove GPL incompatible code
 * Updated command line database upgrade utility
 * Graph zooming now from everywhere

Change Log

 bug#0002228: GPL incompatible files included in Cacti project in
  include/treeview
 bug#0002383: Sanitize the step and id variables CVE-2013-5588,
  CVE-2013-5589
 bug#0002385: Cannot export host templates while including dependencies
 bug#0002386: cli/upgrade_database.php is missing the last two releases
 bug#0002390: Poller/script issue with slash and backslash
 bug#0002405: SQL injection in graph_xport.php
 bug#0002431: CVE-2014-2326 Unspecified HTML Injection Vulnerability
 bug#0002432: CVE-2014-2327 Cross Site Request Forgery Vulnerability -
  Special Thanks to Deutsche Telekom CERT
 bug#0002433: CVE-2014-2328 Unspecified Remote Command Execution
  Vulnerability
 bug#0002434: Suppress SNMP UNITS Suffix from cacti_snmp_get() output
 bug#0002438: Down Host Detection issue when using SNMP Desc or SNMP getNext
 bug#0002446: Subtract plugin processing time from Poller sleep time
 bug#0002453: CVE-2014-4002 Cross-Site Scripting Vulnerability - Special
  Thanks to G. Geshev (munmap)
 bug#0002455: Incomplete and incorrect input parsing leads to remote
  code execution and SQL injection attack scenarios
 bug#0002456: CVE-2014-5025 / CVE-2014-5026 - Cross-Site Scripting
  Vulnerability - Special Thanks to Adan Alvarez and Paul Gevers
 bug: Fix COMMENT handling, even in case COMMENT is empty, with or
  without HR and with variable substitution
 bug: Fix issues when SNMP data holds a "="; "explode" must be treated
  accordingly
 bug: Fix filter highlighting on data sources for the data template field
 bug: correct description of SNMP V3 parameters
 feature: Added native jquery, jqueryui, and jstree
 feature: Fixed issues with 'Clear' under preview not working
 feature: Added new Tree navigation
 feature: Added Columns and Thumbnails to Preview
 feature: Added Columns to Tree (Preview only)
 feature: Both Graphs and Columns default to 'Default'
 feature: Resolved Left hand navigation taking entire page
 feature: Added new graph zoom to tree view and preview offering a
  "quick" (default) and an "advanced" mode

Reporting Bugs

  http://www.cacti.net/bugs.php

Download Cacti

  http://www.cacti.net/download_cacti.php

Download Spine

  http://www.cacti.net/spine_download.php

Thanks!
The Cacti Group

64bits counters are useful when you have interface support them and if you have more 113mbits traffic supported by 32bits snmp counters.

David

> Le 13 nov. 2014 à 20:31, Tom Bilan <tom...@bi...> a écrit :
> 
> Some devices support 64 bit counters.  It's a different OID than normal
> counters.  The 64 bit graph is for devices that have 64 bit counters.
> Bytes means show data in terms of bytes instead of bits.
> 
> Just create one of each and look at how they're different.  If your device
> doesn't support 64 bit counters the graph will be empty.  You can always
> delete the graphs so add one of each type and then standardize on the ones
> you like best.
> 
> On Wed, Nov 12, 2014 at 9:12 PM, per...@gm... <per...@gm...>
> wrote:
> 
>> Hi,
>> At the bottom of the "Create Graphs for this Host" page, there is a
>> “Select a graph type” dropdown box. What is the difference between the
>> three options in the dropdown box: “In/Out Bits (64-bit Counters)”, “In/Out
>> Bits with Total Bandwidth” and “In/Out Bye”?
>> 
>> 
>> 
>> per...@gm...
>> 
>> ------------------------------------------------------------------------------
>> Comprehensive Server Monitoring with Site24x7.
>> Monitor 10 servers for $9/Month.
>> Get alerted through email, SMS, voice calls or mobile push notifications.
>> Take corrective actions from your mobile device.
>> 
>> http://pubads.g.doubleclick.net/gampad/clk?id=154624111&iu=/4140/ostg.clktrk
>> _______________________________________________
>> cacti-user mailing list
>> cac...@li...
>> https://lists.sourceforge.net/lists/listinfo/cacti-user
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for $9/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.
> http://pubads.g.doubleclick.net/gampad/clk?id=154624111&iu=/4140/ostg.clktrk
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user

Some devices support 64 bit counters.  It's a different OID than normal
counters.  The 64 bit graph is for devices that have 64 bit counters.
Bytes means show data in terms of bytes instead of bits.

Just create one of each and look at how they're different.  If your device
doesn't support 64 bit counters the graph will be empty.  You can always
delete the graphs so add one of each type and then standardize on the ones
you like best.

On Wed, Nov 12, 2014 at 9:12 PM, per...@gm... <per...@gm...>
wrote:

> Hi,
> At the bottom of the "Create Graphs for this Host" page, there is a
> “Select a graph type” dropdown box. What is the difference between the
> three options in the dropdown box: “In/Out Bits (64-bit Counters)”, “In/Out
> Bits with Total Bandwidth” and “In/Out Bye”?
>
>
>
> per...@gm...
>
> ------------------------------------------------------------------------------
> Comprehensive Server Monitoring with Site24x7.
> Monitor 10 servers for $9/Month.
> Get alerted through email, SMS, voice calls or mobile push notifications.
> Take corrective actions from your mobile device.
>
> http://pubads.g.doubleclick.net/gampad/clk?id=154624111&iu=/4140/ostg.clktrk
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
>

Hi,
At the bottom of the "Create Graphs for this Host" page, there is a “Select a graph type” dropdown box. What is the difference between the three options in the dropdown box: “In/Out Bits (64-bit Counters)”, “In/Out Bits with Total Bandwidth” and “In/Out Bye”?



per...@gm...

This is definitely an OS or PHP limitation. Per Gandalf, I would lean towards OS, but testing should confirm. 

# xargs --show-limits 
Your environment variables take up 1972 bytes 
POSIX upper limit on argument length (this system): 4611686018427383883 
POSIX smallest allowable upper limit on argument length (all systems): 4096 
Maximum length of command we could actually use: 4611686018427381911 
Size of command buffer we are actually using: 131072 

That's RHEL6, my first guess would be to get the following to increase, afterwards, restart Apache and then go backward from there: 

Size of command buffer we are actually using: 131072 

Regards, 

You know Who 

----- Original Message -----

From: "Paul Gevers" <pa...@cl...> 
To: "Support requests/questions about cacti" <cac...@li...> 
Sent: Monday, October 27, 2014 3:42:35 PM 
Subject: Re: [cacti-user] Issues with Cacti running very long rrdtool commands 

> Does anybody know why a command, while run internal to Cacti, would have a 
> different command line argument character limit than when the command is 
> run at the command prompt by any user on the machine, attempted in various 
> shells? 

No, but I suspect it is due to being called by php. 

> If I could somehow emulate the exact command Cacti is running, with the 
> same environment variables, via the command prompt, that could also help to 
> determine next steps. 

I believe you can run php interactive from the command prompt. That is 
where I would start. 

Paul 


------------------------------------------------------------------------------ 

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

> Does anybody know why a command, while run internal to Cacti, would have a
> different command line argument character limit than when the command is
> run at the command prompt by any user on the machine, attempted in various
> shells?

No, but I suspect it is due to being called by php.

> If I could somehow emulate the exact command Cacti is running, with the
> same environment variables, via the command prompt, that could also help to
> determine next steps.

I believe you can run php interactive from the command prompt. That is
where I would start.

Paul

The suggestion is as follows:
I _may_ happen, that there is a buffer of such size created in the code without
any overflow check. Most likely to be found in lib/rrd.php, but there are other
places I don't remember by heart.
Reinhard

Am 27.10.2014 um 20:34 schrieb Ryan Garland:
> I'll give this one last try, since this apparently affects such a small
> number of Cacti users, and others have been unable to solve this thus far.
> 
> Does anybody know why a command, while run internal to Cacti, would have a
> different command line argument character limit than when the command is
> run at the command prompt by any user on the machine, attempted in various
> shells?
> 
> That is the single question that I believe needs to be answered to help
> pinpoint why this is occurring.  If it can be fixed (i.e. allow Cacti to
> use longer argument strings), I would like to know how, as this would
> likely solve the issue entirely.
> 
> If I could somehow emulate the exact command Cacti is running, with the
> same environment variables, via the command prompt, that could also help to
> determine next steps.
> 
> Any suggestions are appreciated.
> 
> Ryan
> 
> On Thu, Oct 23, 2014 at 11:09 AM, Ryan Garland <sh...@gm...> wrote:
> 
>> If it was an OS limitation, wouldn't the same command fail when run
>> manually at the command line?  Our arg max is quite high.
>>
>> root@netmon100:/usr/local/cacti# getconf ARG_MAX
>> 2097152
>>
>> I found this thread posted by somebody last year describing the exact same
>> issue, but with no resolution:
>>
>> http://forums.cacti.net/viewtopic.php?f=2&t=50123
>>
>> I also confirmed that PHP's memory_limit is listed as 4096M in Cacti
>> (increased from 128M during testing), but that also did not solve the
>> problem.
>>
>> This is fairly easy to reproduce - select 1000 data sources (of the same
>> type) and create an aggregate graph w/ legends using the aggregate plugin
>> and see if the graph draws.  I am curious whether anyone is able to get
>> this to work, and if so, how?
>>
>> If it is not possible to work around this when rrdtool is called via
>> Cacti, we will have to look into other tools to graph this aggregate data
>> and I really don't want to do that.  Cacti has served us very well for many
>> years.
>>
>> Ryan
>>
>> On Thu, Oct 23, 2014 at 10:19 AM, Gandalf <ga...@ca...> wrote:
>>
>>> This is IMHO a restriction of the underlying OS. At least it is known,
>>> that some
>>> of those have a restriction like that
>>> Reinhard
>>>
>>> Am 23.10.2014 um 07:51 schrieb Ryan Garland:
>>>> To follow up on this, I did some tests with the graph that is failing vs
>>>> the graph that is working by adding:
>>>>
>>>> print strlen($command_line);
>>>>
>>>> inside the following function in rrd.php:
>>>>
>>>> function rrdtool_execute
>>>>
>>>> The graph that fails has a string length of 131104, the graph that
>>> succeeds
>>>> has a string length of 130864.  So, it would seem as though we are
>>> running
>>>> into a 131072 limitation.
>>>>
>>>> Any ideas how we can get around this, or get Cacti to run the command
>>> as if
>>>> it were at the command prompt which does not seem to have such a
>>> limitation?
>>>>
>>>> Cheers,
>>>>
>>>> Ryan
>>>>
>>>> On Wed, Oct 22, 2014 at 9:47 PM, Ryan Garland <sh...@gm...> wrote:
>>>>
>>>>> I believe I am running into a character limit for the commands that
>>>>> rrdtool can successfully run.
>>>>>
>>>>> I currently have a stacked aggregate graph with 1376 line items (3/4 of
>>>>> which are the Legend lines).  This graph renders just fine, and with
>>> Debug
>>>>> mode turned on, after the long rrdtool command, rrdtool returns "OK".
>>> The
>>>>> character count of the command (including any white space) of this
>>> command
>>>>> is 132961.  (Yes, I know that's a lot - I have my reasons :))
>>>>>
>>>>> However, as soon as I add a new line item to the graph, the graph no
>>>>> longer renders, and with Debug mode turned on, rrdtool does not return
>>> OK
>>>>> -- it merely shows nothing.  The character count of this command is
>>> 133204.
>>>>>
>>>>> Here's the kicker:  When the failing rrdtool command is manually ran at
>>>>> the system's command prompt, the command succeeds.  It seems that only
>>>>> within Cacti does the rrdtool command not work.
>>>>>
>>>>> I have done a lot of research and found some seemingly related
>>> information
>>>>> ( such as http://www.mail-archive.com/bug...@gn.../msg05522.html )
>>>>> which discusses command length limitations by the OS / kernel,
>>> limiting to
>>>>> 131072 MAX_ARG_STRLEN, but:
>>>>>
>>>>> 1) this number is slightly lower than the length of the full rrdtool
>>>>> command that is succeed, and
>>>>> 2) if that was the problem, the command should not succeed when
>>> manually
>>>>> run at the command prompt
>>>>>
>>>>> So what is Cacti doing when it runs rrdtool that could be limiting the
>>>>> command length?  Does anyone have any suggestions as to how to resolve
>>> this
>>>>> apparent limitation?
>>>>>
>>>>> Any help is appreciated.  Software versions listed below.
>>>>>
>>>>> Cacti Version 0.8.8a
>>>>> PHP Version 5.3.10-1ubuntu3.14
>>>>> Ubuntu Version 12.04
>>>>> RRDTool Version 1.4.x
>>>>>
>>>>> Ryan
>>>>>
>>>>
>>> ------------------------------------------------------------------------------
>>>> _______________________________________________
>>>> cacti-user mailing list
>>>> cac...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/cacti-user
>>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> _______________________________________________
>>> cacti-user mailing list
>>> cac...@li...
>>> https://lists.sourceforge.net/lists/listinfo/cacti-user
>>>
>>
>>
> ------------------------------------------------------------------------------
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
> 

I'll give this one last try, since this apparently affects such a small
number of Cacti users, and others have been unable to solve this thus far.

Does anybody know why a command, while run internal to Cacti, would have a
different command line argument character limit than when the command is
run at the command prompt by any user on the machine, attempted in various
shells?

That is the single question that I believe needs to be answered to help
pinpoint why this is occurring.  If it can be fixed (i.e. allow Cacti to
use longer argument strings), I would like to know how, as this would
likely solve the issue entirely.

If I could somehow emulate the exact command Cacti is running, with the
same environment variables, via the command prompt, that could also help to
determine next steps.

Any suggestions are appreciated.

Ryan

On Thu, Oct 23, 2014 at 11:09 AM, Ryan Garland <sh...@gm...> wrote:

> If it was an OS limitation, wouldn't the same command fail when run
> manually at the command line?  Our arg max is quite high.
>
> root@netmon100:/usr/local/cacti# getconf ARG_MAX
> 2097152
>
> I found this thread posted by somebody last year describing the exact same
> issue, but with no resolution:
>
> http://forums.cacti.net/viewtopic.php?f=2&t=50123
>
> I also confirmed that PHP's memory_limit is listed as 4096M in Cacti
> (increased from 128M during testing), but that also did not solve the
> problem.
>
> This is fairly easy to reproduce - select 1000 data sources (of the same
> type) and create an aggregate graph w/ legends using the aggregate plugin
> and see if the graph draws.  I am curious whether anyone is able to get
> this to work, and if so, how?
>
> If it is not possible to work around this when rrdtool is called via
> Cacti, we will have to look into other tools to graph this aggregate data
> and I really don't want to do that.  Cacti has served us very well for many
> years.
>
> Ryan
>
> On Thu, Oct 23, 2014 at 10:19 AM, Gandalf <ga...@ca...> wrote:
>
>> This is IMHO a restriction of the underlying OS. At least it is known,
>> that some
>> of those have a restriction like that
>> Reinhard
>>
>> Am 23.10.2014 um 07:51 schrieb Ryan Garland:
>> > To follow up on this, I did some tests with the graph that is failing vs
>> > the graph that is working by adding:
>> >
>> > print strlen($command_line);
>> >
>> > inside the following function in rrd.php:
>> >
>> > function rrdtool_execute
>> >
>> > The graph that fails has a string length of 131104, the graph that
>> succeeds
>> > has a string length of 130864.  So, it would seem as though we are
>> running
>> > into a 131072 limitation.
>> >
>> > Any ideas how we can get around this, or get Cacti to run the command
>> as if
>> > it were at the command prompt which does not seem to have such a
>> limitation?
>> >
>> > Cheers,
>> >
>> > Ryan
>> >
>> > On Wed, Oct 22, 2014 at 9:47 PM, Ryan Garland <sh...@gm...> wrote:
>> >
>> >> I believe I am running into a character limit for the commands that
>> >> rrdtool can successfully run.
>> >>
>> >> I currently have a stacked aggregate graph with 1376 line items (3/4 of
>> >> which are the Legend lines).  This graph renders just fine, and with
>> Debug
>> >> mode turned on, after the long rrdtool command, rrdtool returns "OK".
>> The
>> >> character count of the command (including any white space) of this
>> command
>> >> is 132961.  (Yes, I know that's a lot - I have my reasons :))
>> >>
>> >> However, as soon as I add a new line item to the graph, the graph no
>> >> longer renders, and with Debug mode turned on, rrdtool does not return
>> OK
>> >> -- it merely shows nothing.  The character count of this command is
>> 133204.
>> >>
>> >> Here's the kicker:  When the failing rrdtool command is manually ran at
>> >> the system's command prompt, the command succeeds.  It seems that only
>> >> within Cacti does the rrdtool command not work.
>> >>
>> >> I have done a lot of research and found some seemingly related
>> information
>> >> ( such as http://www.mail-archive.com/bug...@gn.../msg05522.html )
>> >> which discusses command length limitations by the OS / kernel,
>> limiting to
>> >> 131072 MAX_ARG_STRLEN, but:
>> >>
>> >> 1) this number is slightly lower than the length of the full rrdtool
>> >> command that is succeed, and
>> >> 2) if that was the problem, the command should not succeed when
>> manually
>> >> run at the command prompt
>> >>
>> >> So what is Cacti doing when it runs rrdtool that could be limiting the
>> >> command length?  Does anyone have any suggestions as to how to resolve
>> this
>> >> apparent limitation?
>> >>
>> >> Any help is appreciated.  Software versions listed below.
>> >>
>> >> Cacti Version 0.8.8a
>> >> PHP Version 5.3.10-1ubuntu3.14
>> >> Ubuntu Version 12.04
>> >> RRDTool Version 1.4.x
>> >>
>> >> Ryan
>> >>
>> >
>> ------------------------------------------------------------------------------
>> > _______________________________________________
>> > cacti-user mailing list
>> > cac...@li...
>> > https://lists.sourceforge.net/lists/listinfo/cacti-user
>> >
>>
>>
>> ------------------------------------------------------------------------------
>> _______________________________________________
>> cacti-user mailing list
>> cac...@li...
>> https://lists.sourceforge.net/lists/listinfo/cacti-user
>>
>
>

If it was an OS limitation, wouldn't the same command fail when run
manually at the command line?  Our arg max is quite high.

root@netmon100:/usr/local/cacti# getconf ARG_MAX
2097152

I found this thread posted by somebody last year describing the exact same
issue, but with no resolution:

http://forums.cacti.net/viewtopic.php?f=2&t=50123

I also confirmed that PHP's memory_limit is listed as 4096M in Cacti
(increased from 128M during testing), but that also did not solve the
problem.

This is fairly easy to reproduce - select 1000 data sources (of the same
type) and create an aggregate graph w/ legends using the aggregate plugin
and see if the graph draws.  I am curious whether anyone is able to get
this to work, and if so, how?

If it is not possible to work around this when rrdtool is called via Cacti,
we will have to look into other tools to graph this aggregate data and I
really don't want to do that.  Cacti has served us very well for many years.

Ryan

On Thu, Oct 23, 2014 at 10:19 AM, Gandalf <ga...@ca...> wrote:

> This is IMHO a restriction of the underlying OS. At least it is known,
> that some
> of those have a restriction like that
> Reinhard
>
> Am 23.10.2014 um 07:51 schrieb Ryan Garland:
> > To follow up on this, I did some tests with the graph that is failing vs
> > the graph that is working by adding:
> >
> > print strlen($command_line);
> >
> > inside the following function in rrd.php:
> >
> > function rrdtool_execute
> >
> > The graph that fails has a string length of 131104, the graph that
> succeeds
> > has a string length of 130864.  So, it would seem as though we are
> running
> > into a 131072 limitation.
> >
> > Any ideas how we can get around this, or get Cacti to run the command as
> if
> > it were at the command prompt which does not seem to have such a
> limitation?
> >
> > Cheers,
> >
> > Ryan
> >
> > On Wed, Oct 22, 2014 at 9:47 PM, Ryan Garland <sh...@gm...> wrote:
> >
> >> I believe I am running into a character limit for the commands that
> >> rrdtool can successfully run.
> >>
> >> I currently have a stacked aggregate graph with 1376 line items (3/4 of
> >> which are the Legend lines).  This graph renders just fine, and with
> Debug
> >> mode turned on, after the long rrdtool command, rrdtool returns "OK".
> The
> >> character count of the command (including any white space) of this
> command
> >> is 132961.  (Yes, I know that's a lot - I have my reasons :))
> >>
> >> However, as soon as I add a new line item to the graph, the graph no
> >> longer renders, and with Debug mode turned on, rrdtool does not return
> OK
> >> -- it merely shows nothing.  The character count of this command is
> 133204.
> >>
> >> Here's the kicker:  When the failing rrdtool command is manually ran at
> >> the system's command prompt, the command succeeds.  It seems that only
> >> within Cacti does the rrdtool command not work.
> >>
> >> I have done a lot of research and found some seemingly related
> information
> >> ( such as http://www.mail-archive.com/bug...@gn.../msg05522.html )
> >> which discusses command length limitations by the OS / kernel, limiting
> to
> >> 131072 MAX_ARG_STRLEN, but:
> >>
> >> 1) this number is slightly lower than the length of the full rrdtool
> >> command that is succeed, and
> >> 2) if that was the problem, the command should not succeed when manually
> >> run at the command prompt
> >>
> >> So what is Cacti doing when it runs rrdtool that could be limiting the
> >> command length?  Does anyone have any suggestions as to how to resolve
> this
> >> apparent limitation?
> >>
> >> Any help is appreciated.  Software versions listed below.
> >>
> >> Cacti Version 0.8.8a
> >> PHP Version 5.3.10-1ubuntu3.14
> >> Ubuntu Version 12.04
> >> RRDTool Version 1.4.x
> >>
> >> Ryan
> >>
> >
> ------------------------------------------------------------------------------
> > _______________________________________________
> > cacti-user mailing list
> > cac...@li...
> > https://lists.sourceforge.net/lists/listinfo/cacti-user
> >
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
>

This is IMHO a restriction of the underlying OS. At least it is known, that some
of those have a restriction like that
Reinhard

Am 23.10.2014 um 07:51 schrieb Ryan Garland:
> To follow up on this, I did some tests with the graph that is failing vs
> the graph that is working by adding:
> 
> print strlen($command_line);
> 
> inside the following function in rrd.php:
> 
> function rrdtool_execute
> 
> The graph that fails has a string length of 131104, the graph that succeeds
> has a string length of 130864.  So, it would seem as though we are running
> into a 131072 limitation.
> 
> Any ideas how we can get around this, or get Cacti to run the command as if
> it were at the command prompt which does not seem to have such a limitation?
> 
> Cheers,
> 
> Ryan
> 
> On Wed, Oct 22, 2014 at 9:47 PM, Ryan Garland <sh...@gm...> wrote:
> 
>> I believe I am running into a character limit for the commands that
>> rrdtool can successfully run.
>>
>> I currently have a stacked aggregate graph with 1376 line items (3/4 of
>> which are the Legend lines).  This graph renders just fine, and with Debug
>> mode turned on, after the long rrdtool command, rrdtool returns "OK".  The
>> character count of the command (including any white space) of this command
>> is 132961.  (Yes, I know that's a lot - I have my reasons :))
>>
>> However, as soon as I add a new line item to the graph, the graph no
>> longer renders, and with Debug mode turned on, rrdtool does not return OK
>> -- it merely shows nothing.  The character count of this command is 133204.
>>
>> Here's the kicker:  When the failing rrdtool command is manually ran at
>> the system's command prompt, the command succeeds.  It seems that only
>> within Cacti does the rrdtool command not work.
>>
>> I have done a lot of research and found some seemingly related information
>> ( such as http://www.mail-archive.com/bug...@gn.../msg05522.html )
>> which discusses command length limitations by the OS / kernel, limiting to
>> 131072 MAX_ARG_STRLEN, but:
>>
>> 1) this number is slightly lower than the length of the full rrdtool
>> command that is succeed, and
>> 2) if that was the problem, the command should not succeed when manually
>> run at the command prompt
>>
>> So what is Cacti doing when it runs rrdtool that could be limiting the
>> command length?  Does anyone have any suggestions as to how to resolve this
>> apparent limitation?
>>
>> Any help is appreciated.  Software versions listed below.
>>
>> Cacti Version 0.8.8a
>> PHP Version 5.3.10-1ubuntu3.14
>> Ubuntu Version 12.04
>> RRDTool Version 1.4.x
>>
>> Ryan
>>
> ------------------------------------------------------------------------------
> _______________________________________________
> cacti-user mailing list
> cac...@li...
> https://lists.sourceforge.net/lists/listinfo/cacti-user
> 

To follow up on this, I did some tests with the graph that is failing vs
the graph that is working by adding:

print strlen($command_line);

inside the following function in rrd.php:

function rrdtool_execute

The graph that fails has a string length of 131104, the graph that succeeds
has a string length of 130864.  So, it would seem as though we are running
into a 131072 limitation.

Any ideas how we can get around this, or get Cacti to run the command as if
it were at the command prompt which does not seem to have such a limitation?

Cheers,

Ryan

On Wed, Oct 22, 2014 at 9:47 PM, Ryan Garland <sh...@gm...> wrote:

> I believe I am running into a character limit for the commands that
> rrdtool can successfully run.
>
> I currently have a stacked aggregate graph with 1376 line items (3/4 of
> which are the Legend lines).  This graph renders just fine, and with Debug
> mode turned on, after the long rrdtool command, rrdtool returns "OK".  The
> character count of the command (including any white space) of this command
> is 132961.  (Yes, I know that's a lot - I have my reasons :))
>
> However, as soon as I add a new line item to the graph, the graph no
> longer renders, and with Debug mode turned on, rrdtool does not return OK
> -- it merely shows nothing.  The character count of this command is 133204.
>
> Here's the kicker:  When the failing rrdtool command is manually ran at
> the system's command prompt, the command succeeds.  It seems that only
> within Cacti does the rrdtool command not work.
>
> I have done a lot of research and found some seemingly related information
> ( such as http://www.mail-archive.com/bug...@gn.../msg05522.html )
> which discusses command length limitations by the OS / kernel, limiting to
> 131072 MAX_ARG_STRLEN, but:
>
> 1) this number is slightly lower than the length of the full rrdtool
> command that is succeed, and
> 2) if that was the problem, the command should not succeed when manually
> run at the command prompt
>
> So what is Cacti doing when it runs rrdtool that could be limiting the
> command length?  Does anyone have any suggestions as to how to resolve this
> apparent limitation?
>
> Any help is appreciated.  Software versions listed below.
>
> Cacti Version 0.8.8a
> PHP Version 5.3.10-1ubuntu3.14
> Ubuntu Version 12.04
> RRDTool Version 1.4.x
>
> Ryan
>

I believe I am running into a character limit for the commands that rrdtool
can successfully run.

I currently have a stacked aggregate graph with 1376 line items (3/4 of
which are the Legend lines).  This graph renders just fine, and with Debug
mode turned on, after the long rrdtool command, rrdtool returns "OK".  The
character count of the command (including any white space) of this command
is 132961.  (Yes, I know that's a lot - I have my reasons :))

However, as soon as I add a new line item to the graph, the graph no longer
renders, and with Debug mode turned on, rrdtool does not return OK -- it
merely shows nothing.  The character count of this command is 133204.

Here's the kicker:  When the failing rrdtool command is manually ran at the
system's command prompt, the command succeeds.  It seems that only within
Cacti does the rrdtool command not work.

I have done a lot of research and found some seemingly related information
( such as http://www.mail-archive.com/bug...@gn.../msg05522.html )
which discusses command length limitations by the OS / kernel, limiting to
131072 MAX_ARG_STRLEN, but:

1) this number is slightly lower than the length of the full rrdtool
command that is succeed, and
2) if that was the problem, the command should not succeed when manually
run at the command prompt

So what is Cacti doing when it runs rrdtool that could be limiting the
command length?  Does anyone have any suggestions as to how to resolve this
apparent limitation?

Any help is appreciated.  Software versions listed below.

Cacti Version 0.8.8a
PHP Version 5.3.10-1ubuntu3.14
Ubuntu Version 12.04
RRDTool Version 1.4.x

Ryan

I've just installed cacti on my debian wheezy system via apt-get.

I managed to get the bundled graphs working though I did have to delete
them and add them again.

Today I installed thold and settings plugins.

For a test, I created a threshold for the number of logged in users, thus I
can log in and get it to trigger the threshold.

I see the threshold go red, but I don't get an email.

However, when I log out, I do get a notification that the threshold went
back to normal, so something seems to be working.

Some other strange thing with thold I noticed, on the graphs, that button
to toggle VHOLD rules, I read in the forum that this is supposed to put a
red line on the graph but I don't see any red line appear.  I did wait
quite a while for the graph to refresh on the server.

I have cacti working flawlessly on a centos 7.0 machine but I need it
working on a centos 6.3 machine.

I download the same version as what is on the 7.0, I installed rrdtool from
yum.

After performing a successful base installation of cacti and confirm it
works. 

I then copy over the database and restart mysqld 

I see my data but 'none of the charts appear' and I get this error in my
cacti.log

 

10/01/2014 04:12:31 AM - CMDPHP: Poller[0] ERROR: SQL Row Failed!,
Error:'1033', SQL:"select  host.id,  host.hostname,  host.snmp_community,
host.snmp_version,  host.snmp_username,  host.snmp_password,
host.snmp_auth_protocol,  host.snmp_priv_passphrase,
host.snmp_priv_protocol,  host.snmp_context,  host.snmp_port,
host.snmp_timeout,  host.disabled  from host  where host.id=1"

 

I have upgraded MySQL to 5.4 and rddtool  to 1.4.8 and tried a different
cacti versions

I have also tried every script in cli/

Mysqlcheck say the database is just fine

Same error

 

 

How do I get my charts to work?

 

 

Brad 

 

Hi cacti-user@,

I'm using cacti with LDAP authentication configured to authenticate with 
an OpenLDAP server.
My ldap user has no right except to bind and search for my own 
attributes in the ldap schema, so no ldap compare works with my own 
user.

In cacti I have set up the "specific DN search" option and provided a 
service account which has more privileges so it can be used to search, 
compare, etc.

The problem i'm facing is that if I check the "require group membership" 
option, cacti is trying to ldap_compare() using a ldap_bind() previously 
done with my credentials instead of the specific DN provided, which in 
my setup fails because my account can't see group attributes in the ldap 
schema.

I would have expected it to use the specific DN to ldap_compare().

Is this done on purpose ? If so, could you help me understand in which 
situation this is usefull ?
I think this can only works with a ldap server who allows your user to 
read group attributes like Microsoft Active Directory.

This patch in Authenticate() from ldap.php works for me:

[remy@cactiserver lib]# diff -u ldap.php.backup_original ldap.php
--- ldap.php.backup_original 2014-08-06 15:27:52.000000000 -0400
+++ ldap.php 2014-08-07 20:21:20.000000000 -0400
@@ -249,6 +249,19 @@
$ldap_response = @ldap_bind($ldap_conn, $this->dn, $this->password);
if ($ldap_response) {
if ($this->group_require == 1) {
+ /* if using specific DN for searching, most likely you want it for 
ldap_compare too */
+ if ($this->mode == "2") {
+ $ldap_response = @ldap_bind($ldap_conn, $this->specific_dn, 
$this->specific_password);
+ if (!$ldap_response) {
+ /* we did this in Search() before, should not happen here */
+ $ldap_error = ldap_errno($ldap_conn);
+ $output["error_num"] = "11";
+ $output["error_text"] = "General bind error, LDAP result: " . 
ldap_error($ldap_conn);
+ cacti_log("LDAP: " . $output["error_text"], false, "BIND");
+ @ldap_close($ldap_conn);
+ return $output;
+ }
+ }
/* Process group membership if required */
if ($this->group_member_type == 1) {
$ldap_group_response = @ldap_compare($ldap_conn, $this->group_dn, 
$this->group_attrib, $this->dn);

Thank you

Hi Mark

On 27-08-14 16:11, Mark Haney wrote:
> Got it.  It was the 0.8.8b COMMENT bug.  Has that bug not been posted to
> the 0.8.8b tarball?  I mean, it's been out nearly a year and the patch I
> used was posted in March.

That would have been 0.8.8c and no, that is not released yet.

Paul

Got it.  It was the 0.8.8b COMMENT bug.  Has that bug not been posted to
the 0.8.8b tarball?  I mean, it's been out nearly a year and the patch I
used was posted in March.


On Wed, Aug 27, 2014 at 9:43 AM, Mark Haney <mar...@vi...>
wrote:

> I think I figured out the problem.  It looks like it's because I'm trying
> to graph a VLAN interface, not a standard interface.
>
>
> On Mon, Aug 25, 2014 at 10:34 AM, Lojze Blatnik <loj...@gm...>
> wrote:
>
>> Again, set cacti log level on medium or higher and you'll see what is
>> going
>> wrong.
>> You can filter them with Host[id] and you'll get just data of cisco
>> switch.
>> On cacti website are a lot of debugging approaches which can help you.
>>
>> BR
>>
>>
>> On 25 August 2014 15:41, Mark Haney <mar...@vi...> wrote:
>>
>> > I found part of my problem.  The cacti user account had an incorrect
>> home
>> > directory. I had originally installed the RPM,but removed it once I
>> > realized the RPM dumps files all over the place.  I didn't catch the
>> cacti
>> > user had been created by the RPM and set the home directory to
>> > /usr/share/cacti.  Now that works, and I'm getting localhost data
>> graphed.
>> > However, I can't get the one switch I've added to cacti to graph. I've
>> > added SNMP to firewalld do to pass the data, Cacti can see the Switch
>> (it's
>> > a Cisco Catalyst) and the interfaces on it, but I'm not getting any
>> graphs.
>> >
>> >
>> > It's been a while since I've messed with snmp, so where do I start?
>> >
>> >
>> >
>> > On Mon, Aug 25, 2014 at 8:50 AM, Lojze Blatnik <loj...@gm...
>> >
>> > wrote:
>> >
>> > > turn cacti log level to medium and you'll see if there are any events
>> in
>> > > cacti log file.
>> > > You can alternatively do following hack if cron is really the issue.
>> > > yum install screen
>> > >
>> > > screen -dm -S cacti_poller  bash -c "for ((;;)); do php
>> poller.php;sleep
>> > > 5m; done"
>> > >
>> > > ... to see the terminal
>> > > screen -r
>> > >
>> > > .. to put it in background again
>> > > ctrl + a <release for half of second> d
>> > >
>> > > Good luck
>> > >
>> > >
>> > > On 25 August 2014 13:28, Mark Haney <mar...@vi...>
>> wrote:
>> > >
>> > > > The timezone settings are correct.  It looks as if the cron job
>> isn't
>> > > > running, however.  I have tried it as root and as the user 'cacti'
>> > which
>> > > I
>> > > > setup when installing it.  It's very strange.
>> > > >
>> > > >
>> > > >
>> > > > On Fri, Aug 22, 2014 at 3:08 PM, Paul Gevers <pa...@cl...>
>> wrote:
>> > > >
>> > > > > Check your timezone settings. It is in three places, system, mysql
>> > and
>> > > > > php. They need to be set to the same thing.
>> > > > >
>> > > > > Paul
>> > > > >
>> > > > > On 22-08-14 20:07, Mark Haney wrote:
>> > > > > > I've got Cacti setup on CentOS 7, but I'm encountering a problem
>> > > > getting
>> > > > > > any graph data. I get graphs, just nothing in them, including
>> the
>> > > > > > localhost graphs. I've setup the file permissions on the rra/
>> and
>> > > log/
>> > > > > > directories to the 'cacti' user and the poller creates and
>> appears
>> > to
>> > > > > > update the RRD files, but still no data in the graphs.
>> > > > > >
>> > > > > > I've disabled SELinux, but that hasn't done anything to fix it.
>> > > Anyone
>> > > > > > got any ideas on where I look next?
>> > > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > > >
>> > > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > > > Slashdot TV.
>> > > > > Video for Nerds.  Stuff that matters.
>> > > > > http://tv.slashdot.org/
>> > > > > _______________________________________________
>> > > > > cacti-user mailing list
>> > > > > cac...@li...
>> > > > > https://lists.sourceforge.net/lists/listinfo/cacti-user
>> > > > >
>> > > > >
>> > > >
>> > > >
>> > > > --
>> > > > Mark Haney
>> > > > Sr. Systems Engineer, VIF International Education
>> > > > mar...@vi...
>> > > > 919-265-5006
>> > > >
>> > > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > > Slashdot TV.
>> > > > Video for Nerds.  Stuff that matters.
>> > > > http://tv.slashdot.org/
>> > > > _______________________________________________
>> > > > cacti-user mailing list
>> > > > cac...@li...
>> > > > https://lists.sourceforge.net/lists/listinfo/cacti-user
>> > > >
>> > >
>> > >
>> >
>> ------------------------------------------------------------------------------
>> > > Slashdot TV.
>> > > Video for Nerds.  Stuff that matters.
>> > > http://tv.slashdot.org/
>> > > _______________________________________________
>> > > cacti-user mailing list
>> > > cac...@li...
>> > > https://lists.sourceforge.net/lists/listinfo/cacti-user
>> > >
>> >
>> >
>> >
>> > --
>> > Mark Haney
>> > Sr. Systems Engineer, VIF International Education
>> > mar...@vi...
>> > 919-265-5006
>> >
>> >
>> ------------------------------------------------------------------------------
>> > Slashdot TV.
>> > Video for Nerds.  Stuff that matters.
>> > http://tv.slashdot.org/
>> > _______________________________________________
>> > cacti-user mailing list
>> > cac...@li...
>> > https://lists.sourceforge.net/lists/listinfo/cacti-user
>> >
>>
>> ------------------------------------------------------------------------------
>> Slashdot TV.
>> Video for Nerds.  Stuff that matters.
>> http://tv.slashdot.org/
>> _______________________________________________
>> cacti-user mailing list
>> cac...@li...
>> https://lists.sourceforge.net/lists/listinfo/cacti-user
>>
>
>
>
> --
> Mark Haney
> Sr. Systems Engineer, VIF International Education
> mar...@vi...
> 919-265-5006
>



-- 
Mark Haney
Sr. Systems Engineer, VIF International Education
mar...@vi...
919-265-5006