Cacti

Has anyone successfully upgraded from Cacti with MySQL 4 to MySQL 5 on an existing Cacti install? I have another monitoring app that I'd like to run which now requires MySQL 5 (Groundwork Monitor), but I don't know if migrating Cacti to MySQL 5 will work, or if I'll just break everything. I do see t...

Will you make this available for those of us running 0.86g? Or are there changes we could make to get it to run without having a machine already with 086f?

Roger

Someone had turned full log debugging on without bothering to tell anyone. Log had filled up to 2gb filesize.

<changing the Cacti admin password to something else>

Roger

This is a new problem. It started happening prior to 6am today. All graphs stopped working, and my root email on the box is full of errors like this: Subject: Cron <cacti@monitor2> php /var/www/html/cacti/poller.php > /dev/null 2>&1 X-Cron-Env: <SHELL=/bin/bash> X-Cron-Env: <PATH=/sbin:/bin:/usr...

I've seen references to this elsewhere in the forums, but I need a definite answer. Are all graphs Cacti generates backwards from how we would imagine the data flowing? In other words, looking at a bandwidth graph for a DS3 on a Cisco router, which carries all of our internet traffic, the data to me...

Any idea about why this is happening? I'm using your the from just above. I don't understand why there are parts of data and then red. I've had this running for several days now. The second image is from the smokeping script in cacti. By comparison, the first image is from my live smokeping install....

Well, this script returns accurate results, and on a single line. So that's a huge improvement. And running a diff between it and the script back a ways in this thread shows tiny tiny differences.

Don'tcha just love scripting?

Thanks. Now I've got to get cacti to graph the output.

Roger

I'm using the fping script from earlier in this thread, and I get # perl /var/www/html/cacti/scripts/ping-latency.pl 66.121.1.37 66.121.1.37 : xmt/rcv/%loss = 10/10/0%, min/avg/max = 23.4/63.4/105 min:0.000 avg:0.000 max:0.000 dev:0.000 loss:0 This is on a Fedora Core box. I know fping works okay be...

Hmm.

Does that mean the red line is just broken? Or am I missing something to get 95 percentile working?

Roger

First I tried just changing the graph template. That didn't even provide the 95 percentile data - the data area just had zero. So I deleted the graph and created a new one using the 95 percentile template. In retrospect I did not delete the data source that the graph is drawing from. Do I need to de...

I have been graphing traffic to a specific host for a while with, just using the in/out bits graph template. Converted it to using the in/out bits w/95 percentile template today. It's obviously making a calculation, because I'm getting 2.76 mbit in+out on the graph, but the red line is staying at ze...

Okay, the 95 percentile is an internal cacti function. But it graphs it using an HRULE, and other posts here (search for CIR) have said that using an HRULE in a graph is a good way to graph CIR. Let me explain this way: I need to specify a max value for data traffic on a bits in/out interface, and h...

Here are the graph and data templates I created for use with the hiperdsp.pl script. It looks like you can't export a data input method or I would export that as well, but I don't think that's where the problem lies.

Roger

Someone out there has to know how to do this, I can't be the only one who has read the documentation and found zero help on this topic.

It would be really nice if we could get the full smokeping functionality here too, using fping instead of regular ping. fping gives you a much more accurate picture of latency.