Ad blocker detected: Our website is made possible by displaying online advertisements to our visitors. Please consider supporting us by disabling your ad blocker on our website.
I'm new to this tool and am trying to set it up on a MDK9.0 webserver.
Installation went a breeze and graphs are coming thick and fast.
I have three other Linux boxes and five Solaris (2. boxes which I have set up as polling hosts and I get graphs from all of them... but they all look very similar (not identical just similar) that I don't believe that it is actually working. e.g. Logged in users is 10 on them all, No of Processes is approx the same on a Linux box as the Sun server - which is NOT correct.
Is there something (daemon) that should be running on each host for cacti to get information from? rrdtools and net-snmp are installed on all machines.
TIA,
Steve
Can no-one help with this or am I just being dumb?
How does everyone else run cacti to collate data from remote workstations?
Do you run rrdtool on the remote node and nfs or ftp the data back to the web server.
I started snmp running on one of the Linux boxes (not the web server) but I still get the same results ie the data over time is the same as the web server data even though they are sourcing different rrd files from the rra directory.
If you go to "Graph Management" and select a graph, make sure the selections under "Graph Item Inputs" aren't the same for each graph. If so, that is the problem.
I'll try to keep informations of 3 linux and 3 unix machines.
The graphs looks also very very similar, and I don't know where my problem is.
On the "remote machines" snmpd is running.
I choose "Polling Hosts" from the menu on the left, filled in the Description, the hostname, the Management-IP Address and the SNMP Community (public).
Choose "Local Linux Machine" in the Host Template section and the graphs look very similar. Tried also "Generic SNMP Host and None" but the graphs also look very similar.
I think they are the same as the Localhost Graphs.
If I go to Graph Management and select graph, the selections under "Graph Item Inputs" aren't the same, got there:
eMailserver Load Average (1min)
eMailserver Load Average (15min)
eMailserver Load Average (5min)
and
Localhost Load Average (1min)
Localhost Load Average (15min)
Localhost Load Average (5min)
and so on.
I get the same results as himo re Graph Item Inputs.
I've also tried Local Linux Machine and Generic SNMP Host, and created new templates for HPUX and Solaris.
Only one of my other machines is running snmpd but its graphs look the same as all the rest.
Presumably snmpd does not _need_ to be running for data/graphs to be produced?!
Notice: Undefined variable: regexps in /var/www/html/cacti-0.8.3a/include/rrd_functions.php on line 296
Notice: Undefined variable: fetch_array in /var/www/html/cacti-0.8.3a/include/rrd_functions.php on line 309
ERROR: no rrdtool fetch output, you probably forgot to pass a valid data source
Notice: Undefined variable: regexps in /var/www/html/cacti-0.8.3a/include/rrd_functions.php on line 296
Notice: Undefined variable: fetch_array in /var/www/html/cacti-0.8.3a/include/rrd_functions.php on line 309
ERROR: no rrdtool fetch output, you probably forgot to pass a valid data source
Anonymous wrote:When I try to create a trffic graph I get;
Notice: Undefined variable: regexps in /var/www/html/cacti-0.8.3a/include/rrd_functions.php on line 296
Notice: Undefined variable: fetch_array in /var/www/html/cacti-0.8.3a/include/rrd_functions.php on line 309
ERROR: no rrdtool fetch output, you probably forgot to pass a valid data source
Notice: Undefined variable: regexps in /var/www/html/cacti-0.8.3a/include/rrd_functions.php on line 296
Notice: Undefined variable: fetch_array in /var/www/html/cacti-0.8.3a/include/rrd_functions.php on line 309
ERROR: no rrdtool fetch output, you probably forgot to pass a valid data source
You are getting all of these messages because this graph does not have any data sources assigned to it. If you goto "Graph Management" and select this graph, you should see two dropdowns under the "Graph Item Inputs" box. Even if there are already items selected here, click "save" on this page. It should clear up the graph errors that you are seeing.
I'm having the same problem described above with my linux servers only, all graph looks very similar to the localhost. I tried to generate traffic graph but can't get nothing (no errors, no graph -> the graph is empty). I don't know is there is a problem with snmp or something
If you are looking for a quick fix, I would switch to a graph template that does not use 95th percentile or bandwidth summation. The "Interface - Traffic (bits/sec)" would be a good template to test with. I am not guaranteeing that switching templates will completely solve your problem, but it will at minimum change the error message .
Do you still have any items on your graph that look like "|95:bits:0:max:2|"? Other than the suggestion I made two posts back, I am sort of running out of ideas :-/.
boxes which I have set up as polling hosts and I get graphs from all of them... but they all look very similar (not identical just similar) that I don't believe that it is actually working. e.g. Logged in users is 10 on them all, No of Processes is approx the same on a Linux box as the Sun server - which is NOT correct.
.