[FIX] thold 0.4.3-2

[sodium]

remove the last / from the settings at the base url in the settings from threshold.

check for the test the "Send alerts as text" and see what you get

(you could check if this works with a stock php.ini)

and update to 0.4.4-41 (and run upgrade.php)

[CanDee]

still

Could not open!
http://xxx.xxx.xxx/cacti/graph_image.ph ... _type=tree

i i activate this threshold i get tons of sql errors in cacti.log:

02/25/2011 02:58:12 PM - CMDPHP: Poller[0] ERROR: SQL Row Failed!, Error:'1064', SQL:"select * from host where id="
02/25/2011 02:58:12 PM - CMDPHP: Poller[0] ERROR: SQL Row Failed!, Error:'1064', SQL:"select * from host where id="
02/25/2011 02:58:12 PM - CMDPHP: Poller[0] ERROR: SQL Row Failed!, Error:'1064', SQL:"select * from host where id="

[sodium]

what did you do all of the questions or some?

[CanDee]

i removed the ending / in settings base url
i did php upgrade.php

it works for all other graphs. for example load or cpu usage...
email will be send containing the graph.

only this cacti poller thing shows these errors

[sodium]

there must be something wrong then with the database is all other graphs work fine...


check all parameters of the templates (data & graph) and recreate the graph...

[hmorandell]

i removed the ending / in settings base url
i did php upgrade.php

it works for all other graphs. for example load or cpu usage...
email will be send containing the graph.

only this cacti poller thing shows these errors

exactly same problem for me!
it's a bit annoying, but nothing really bad. everything else works fine for me....

In a post earlier in this thread by dragossto there was a README.txt included which talked about version 0.50 of thold with the following changelog

----[ Changelog

--- 0.5 ---
Bug: Fix for compatibility with other plugins using datasource action hook
Bug: Re-add syslog messages for down hosts
Bug: Fixed a few minor issues
Bug: Allow the use of query_XYZ in CDEFs
Bug: Fix host status page to only allow users to see hosts they have access to
Bug: Fix ru_nswap errors on Windows
Bug: Actions did not work with other plugins from Graphs and Data Sources pages
Feature: Add support for RPN Expressions
Feature: Add support for Alert Escalation (Multiple Alerts per Threshold)
Feature: Add support for Alerts via Script Execution
Feature: Add support for Alerts via SNMP Writes
Feature: Sort current values numerically on THold tab

Any idea which version of thold is currently the most recent????? I'm getting very confused now, who is the official developer/responsible for making releases?

[dragossto]

the latest version from svn it 0.4.3
http://svn.cacti.net/viewvc/thold/branc ... ot=Plugins
whit one year early was 0.5 , but after a while rename whit 0.4.2

whit thold i don't have any problem
check the installation of PIA
in config.php
if you have $url_path = "/cacti/";
put the correct path http://yourip/cacti/
if this setting it not in the config.php
look in global.php after this $url_path = "" an configure whit your path
in $url_path put only the path after your base address(ip)
after this setting made properly you should not have any problem whit plugins

[CanDee]

setting the url_path in global.php has no effect for me...

still recieve the thold email

Could not open!
h**p://my.cactihost/cacti/graph_image.php?local_graph_id=0&rra_id=0&view_type=tree

when i browse to the graph
h**p://my.cactihost/cacti/graph.php?action=view&rra_id=all&local_graph_id=6102


but as i mentioned...only for this graph. (cacti statistics)
all other hosts/graph work fine..

[dragossto]

h**p://my.cactihost/cacti/graph_image.php?local_graph_id=0&rra_id=0&view_type=tree

some thing it wrong hear
the graph id it 0 , it not possible

it possible to make one screen shot ,because for the moment i don't understand
about what statistic you talk about

Have a nice day

[rmcdonald]

So I too like many have this issue with the latest thold version
.4.4.41
PIA 2.9
settings.5

Nothing is triggering I tried downgrading back to .4.2 but now that wornt work has anyone figured this out yet?

Thanks

[hmorandell]

So I too like many have this issue with the latest thold version
.4.4.41
PIA 2.9
settings.5

Nothing is triggering I tried downgrading back to .4.2 but now that wornt work has anyone figured this out yet?

Thanks

well. I solved this by cancelling thold's for the localhost cacti statistics.... everything else works fine for me.
so I do not touch my (currently) good working system

[rmcdonald]

I went back to .4.2 and it is working agaion sigh.

[Velnias]

I have the same problem with my own created graphs and graph templates.
When i create thold for one of them, i got the the link in the mail with graph_id=0.

URL: http: //monitor/cacti//graph.php?local_graph_id=0&rra_id=1

So i went to cacti DB, and in thold_data table found that graph_id and graph_template were 0. so put manually the correct values, and now it works.
Using thold v0.4.3

[deu439356]

What types of RRA settings work the best for baseline? I wanted to use something like a week for reference, but I am not sure how to go about that. Using my current settings, I have a 30 minute or one hour average for my baseline. I know that either of those is not long enough for a good baseline. Here are my two options that I'd like to change:

Code: Select all

                     Step         Rows           Timespan
30 minutes       6 	     8760 	       7884000 
60 minutes       12 	     8760 	       31536000 

Any help or suggestions would be appreciated.

Thanks!

[sodium]

What types of RRA settings work the best for baseline? I wanted to use something like a week for reference, but I am not sure how to go about that.

depends

the used thold version (latest versions use true baseline)

the reference value depends on the data in the graph -- if you have a steady value you can use 30 mins
the longer ago the reference value -> the lower the chance you get a false positive
you mileage may vary...