Re: Problems importing BIND9 SNMP template cof cacti

[Sorry for forwarding this privatly sent mail to a public list, but I hate repeating myself :)]

Quoting Kevin Myer <kevin_myer <at> iu13.org>:

> I'm seeing a lot of Unmet Dependencies when I try to import any of the bind9 XML
> templates into Cacti.  Is this a known issue?

Yes. See http://sourceforge.net/mailarchive/message.php?msg_id=11634191

> Also, in the Readme.txt, it states that "rndc status" is called.  I believe that
> should be "rndc stats" instead.

'status' work for me. What version of bind 9 do you use? I'm using 9.2.5...

>    + Unmet Dependency: (Data Input Method) Get SNMP Data (Indexed)
>    + Unmet Dependency: (Data Input Field) Output Type ID
>    + Unmet Dependency: (Data Input Field) Index Value
>    + Unmet Dependency: (Data Input Field) Index Type
>    + Unmet Dependency: (Data Input Field) SNMP Version (1, 2, or 3)
>    + Unmet Dependency: (Data Input Field) SNMP Password (v3)
>    + Unmet Dependency: (Data Input Field) SNMP Username (v3)
>    + Unmet Dependency: (Data Input Field) SNMP Community
>    + Unmet Dependency: (Data Input Field) SNMP IP Address
>    + Unmet Dependency: (GPRINT Preset) Normal
>    + Unmet Dependency: (Data Input Method) Get SNMP Data (Indexed)

The problem (for me) is that all these IS there, but with a different
hash value...

I've extracted all these dependencies (included). If you dare, you're welcome

Re: Graphing Bind 9 data using SNMP...

Quoting Will McDonald <wmcdonald <at> gmail.com>:

> On 6/22/05, Turbo Fredriksson <turbo <at> bayour.com> wrote:
>> Quoting Will McDonald <wmcdonald <at> gmail.com>:
>> 
>> > I've copied the XML declaration into ~cacti/snmp_queries and imported
>> > cacti_data_query_snmp_local_bind9_statistics.xml, but at this stage I
>> > received a few errors...
>> > [...]
>> > I now have "SNMP - Local - Bind9 Statistics" in Graph Templates in
>> > Cacti, "SNMP - Local - Bind9 Statistics" is listed in Data Templates
>> > but has a Data Input Method of "None", should that be Get SNMP Data
>> > (Indexed)?
>> 
>> This have been discussed a couple of times on the list...
>> 
>> Look at this URL (so I don't have to repeat my self :)
>> 
>> http://sourceforge.net/mailarchive/message.php?msg_id=11634191
>
> thanks for that. I'm retrying this from scratch with the 1.1 tarball
> and I still don't quite get it though. What do I need to do to import
> these XML templates and fulfill these dependancies?
>
> I assume I need to get some hashes out of templates locally and place
> them into cacti_data_query_snmp_local_bind9_statistics_domains.xml and
> cacti_data_query_snmp_local_bind9_statistics_totals.xml? But which
> hashes from where, and where do they go?

Have a look at http://sourceforge.net/mailarchive/forum.php?thread_id=7632776&forum_id=12795

Re: Re: Problems importing BIND9 SNMP template cof cacti

On 01/07/05, Turbo Fredriksson <turbo <at> bayour.com> wrote:
> [Sorry for forwarding this privatly sent mail to a public list, but I hate repeating myself :)]

> Example:
> In the included file, the 'Output Type ID' is specified with a hash
> value of '070006e6deda7be0f391399c5130e7c4a48b28'. This value is specified in
> one place each in the 'cacti_data_query_snmp_local_bind9_statistics_{domains,totals}.xml'
> files:
> 
> So... If you extract something like 'Data Query->Karlnet - Wireless Bridge Statistics'
> in YOUR Cacti installation, you'll get YOUR hash value for the missing dependencies...
> Be sure to tick the 'Include Dependencies' box!

Excellent, I appear to be getting somewhere with this additional info.

I exported Wireless Bridge Statistics and the hash values for the
missing dependencies in this exported XML are identical to the ones in
the previously attached
cacti_data_query_snmp_local_bind9_statistics_dependencies.xml

Importing cacti_data_query_snmp_local_bind9_statistics_dependencies.xml
then cacti_data_query_snmp_local_bind9_statistics_totals.xml still
results in those unmet dependencies. Checking these unment
dependencies the hashes are in both Wireless Bridge Statistics and
cacti_data_query_snmp_local_bind9_statistics_dependencies.xml but
aren't included in
cacti_data_query_snmp_local_bind9_statistics_totals.xml. Which makes
sense as this was exported without its dependencies.

Adding the contents of

Re: Re: Problems importing BIND9 SNMP template cof cacti

> [Sorry for forwarding this privatly sent mail to a public list, but I hate
repeating myself
:)]

Not a problem, except I wasn't subscribed to cacti-users (I am now).  So I never
saw your response until I was CCed on Will's response.

>> Also, in the Readme.txt, it states that "rndc status" is called.  I believe
that
>> should be "rndc stats" instead.
>
> "status" work for me. What version of bind 9 do you use? I"m using 9.2.5...

I'm using 9.2.4.  "rndc status" gives a status report, "rndc stats" dumps the
stats to the stats filed specified in named.conf, for me.

Regarding the need to merge dependencies based on what your local hash values
are - that is NOT user-friendly at all.  Hard coding dependencies in an XML
file is certainly not the way to go.  There must be some way to have cacti
resolve those dependencies to the local hash values, when the items are
imported.  But I'll merge them manually and see what happens.

--

-- 
Kevin M. Myer
Senior Systems Administrator
Lancaster-Lebanon Intermediate Unit 13  http://www.iu13.org

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,

Intermittent graphing problems with 0.86e

Hello,

I've seen this discussed on this list and the forums, particularly related to a
couple of bugs in 0.86d.  However, I upgraded to 0.86e and I'm still seeing the
intermittent graphing problem.  I have a total of 4 polled hosts and the poller
takes less than 2 seconds to run normally.

I am using poller.php from cron every five minutes, although I also tried cactid
(and that wouldn't create any RRAs).  Logs are full of:

07/01/2005 05:00:02 PM - CMDPHP: Poller[0] ERROR: Either there are no items in
the cache or polling is disabled
07/01/2005 05:04:58 PM - POLLER: Poller[0] Maximum runtime of 296 seconds
exceeded. Exiting.

When I run poller from the command line, it completes successfully:

07/01/2005 05:09:39 PM - SYSTEM STATS: Time: 1.2253 s, Method: cmd.php,
Processes: 10, Threads: N/A, Hosts: 6, Hosts/Process: 1, Data Sources 198, RRDs
Processed 70

Although there are a few errors:
ERROR: illegal attempt to update using time 1120252178 when last update time is
1120252178 (minimum one second step)

I can at least get trends from the intermittent data, and for the year views,
its smoothed out :)  But I'd like to track down the cause of the gaps and
eliminate them.

OS is Red Hat Enterprise Linux 3, with the PHP package from RH EL 4 (PHP

95-percentile on data aggregation

Hi there,

I've spent quite a few hours tearing my hair out over this one, looking
through the forums and mail list archives, for a solution to what would
appear to be a trivial problem.

Let's say that I have 2 (or more) circuits with one transit provider.  I
can easily generate graphs to produce an aggregate graph summing up the
input/output over those links, I can also (using the builtin templates),
generate graphs with 95-pile figures for each circuit.

However, with my simple aggregate graph, it appears that the |95:...|
function is not being applied to the aggregate input/output, rather just
the last dataset of the aggregate.

So, I lay my head before the mercy of the assembled masses to ask what
*must* be an FAQ, but I cannot for the life of me find the answer :)

Given two input sources (for the sake of argument, let's make it easy
and call them Cisco interfaces), what magick must I incant in order to
get a 95-pile figure based on the aggregate of the input/output across
those two sources?

Cheers,
	-aDe

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to

(no subject)

Re: (no subject)

Quoting larryjadams <at> comcast.net:

> Kevin,
>
> The error message below is typically found when you have two cron 
> jobs running.  However, I have also found this issue if you are beta 
> testing php 5.1b2.  So, please refrain from using it for now.  Since 
> you are using 4.3.9, you should review your crons.

I thought of that after I sent the message.  I probably touched off a 
manual job
at the same time that the cron job was running.  There is definitely only one
cron job configured, so I either caught that one that runs for a couple of
seconds, or it was in one of its states where it runs for 296 seconds.

That explains one set of message but that still does not solve the gaps 
problem.
I can run the poller from the command line and it takes 1.2 seconds to
complete.  From cron, it logs errors that the alloted time of 296 
seconds is up
and that's always preceded by:

07/02/2005 07:35:00 AM - CMDPHP: Poller[0] ERROR: Either there are no items in
the cache or polling is disabled

And there are these:

07/02/2005 06:15:02 AM - CMDPHP: Poller[0] ASSERT: '37192383<37192383' failed.
Recaching host 'pyroxene.iu13.org', data query #2.

Even when the poller does successfully complete, there's still the 
message about
no items in the cache or polling is disabled:

07/02/2005 02:35:01 AM - CMDPHP: Poller[0] ERROR: Either there are no items in
the cache or polling is disabled
07/02/2005 02:35:02 AM - SYSTEM STATS: Time: 1.7862 s, Method: cmd.php,
Processes: 10, Threads: N/A, Hosts: 6, Hosts/Process: 1, Data Sources 
198, RRDs
Processed 40

So either I have to assume that ERROR is meaningless (in which case it 
probably
shouldn't be logged as an error), or there is something wrong somewhere.

Kevin

--

-- 
Kevin M. Myer
Senior Systems Administrator
Lancaster-Lebanon Intermediate Unit 13  http://www.iu13.org

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click

RE: 95-percentile on data aggregation

Ade,

Yes, there is a bug in the 95th calculation for aggregated values.  I
submitted the bug report on this issue. See below for the bug report.

Your aggregated "total" 95th appears wrong because it is actually 95th(ds0)
+ 95th(ds1) instead of 95th(ds0+ds1). I have confirmed this. While usually
close, you will see some graphs where the aggregate 95th can actually exceed
the max reported on the same data.

Dave

Bug re
The following issue has been ASSIGNED.
======================================================================
http://bugs.cacti.net/view.php?id=490
======================================================================
Reported By:                David Williams
Assigned To:                iberry <at> raxnet.net
======================================================================
Project:                    Cacti
Issue ID:                   490
Category:                   Templates (Graph, Data, Host)
Reproducibility:            always
Severity:                   major
Priority:                   normal
Status:                     assigned
======================================================================
Date Submitted:             06-20-2005 12:30 EDT
Last Modified:              07-01-2005 14:42 EDT
======================================================================
Summary:                    95th Percentile Total is calculated incorrectly
Description:
A graph template item using the value |95:bits:0:total:2| will generate a
95th percentile that is incorrect and potentially larger than the maximum
of the highest in+out on the graph.

Cacti is summing the ds0 95th and ds1 95th, instead of adding ds0+ds1,
then taking the 95th of that series of results.
======================================================================

Issue History
Date Modified   Username       Field                    Change
======================================================================
06-20-05 12:30  David Williams New Issue
07-01-05 14:42  TheWitness     Status                   new => assigned
07-01-05 14:42  TheWitness     Assigned To               =>
iberry <at> raxnet.net
======================================================================

-----Original Message-----
From: cacti-user-admin <at> lists.sourceforge.net
[mailto:cacti-user-admin <at> lists.sourceforge.net]On Behalf Of Ade Lovett
Sent: Saturday, July 02, 2005 12:44 AM
To: cacti-user <at> lists.sourceforge.net
Subject: [cacti-user] 95-percentile on data aggregation

Hi there,

I've spent quite a few hours tearing my hair out over this one, looking
through the forums and mail list archives, for a solution to what would
appear to be a trivial problem.

Let's say that I have 2 (or more) circuits with one transit provider.  I
can easily generate graphs to produce an aggregate graph summing up the
input/output over those links, I can also (using the builtin templates),
generate graphs with 95-pile figures for each circuit.

However, with my simple aggregate graph, it appears that the |95:...|
function is not being applied to the aggregate input/output, rather just
the last dataset of the aggregate.

So, I lay my head before the mercy of the assembled masses to ask what
*must* be an FAQ, but I cannot for the life of me find the answer :)

Given two input sources (for the sake of argument, let's make it easy
and call them Cisco interfaces), what magick must I incant in order to
get a 95-pile figure based on the aggregate of the input/output across
those two sources?

Cheers,
	-aDe

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
cacti-user mailing list
cacti-user <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/cacti-user

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click

Re: (no subject)

Quoting Kevin Myer <kevin_myer <at> iu13.org>:

> Quoting larryjadams <at> comcast.net:
>
>> Kevin,
>>
>> The error message below is typically found when you have two cron 
>> jobs running.  However, I have also found this issue if you are beta 
>> testing php 5.1b2.  So, please refrain from using it for now.  Since 
>> you are using 4.3.9, you should review your crons.
>
> I thought of that after I sent the message.  I probably touched off a 
> manual job
> at the same time that the cron job was running.  There is definitely only one
> cron job configured, so I either caught that one that runs for a couple of
> seconds, or it was in one of its states where it runs for 296 seconds.

I'll eat my words on this after all.  After further investigation, the 
cacti RPM
installation includes a cronjob in /etc/cron.d/cacti.  I discovered this after
I disabled the cronjob I had setup in the crontab of another account, 
that runs
the polling, and found that poller.php was still running every five minutes.

So for RPM users, don't follow the instructions in the documentation.  The
cronjob is already set up.  And as a suggestion, it may be wise to remove the
inclusion of that cronjob from the RPM.

Kevin

--

-- 
Kevin M. Myer
Senior Systems Administrator
Lancaster-Lebanon Intermediate Unit 13  http://www.iu13.org

-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click