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 running 0.8.2 as of yesterday. Fantastic work! Thanks alot!
A problem I noticed is that after adding two LINE1s to the Traffic (bytes/sec) template with MAX instead of AVERAGE (to produce MRTG-like graphs), the Traffic graphs stop working. When I run the output of the 'source' link on such a graph. RRDTOOL gives me:
Instead of adding two LINE1 lines, simply changing the standard 'Interface - Traffic (bytes/sec) graph' AVERAGE CF types for Inbound and Outbound to MAX break stuff. Compare:
What you are trying to do this very possible. What you have done so far is correct, but I think you missed a step. After adding the two LINE1 items to the graph template, you must include these items in the "Graph Item Inputs". The "Inbound Data Source" input will include the MAX-IN item, and the "Outbound Data Source" will include the MAX-OUT item.
Let me know if you have any more problems or questions.
I followed your advice (which is obvious - sorry for missing that), but now something really odd has happened: all the DEF: lines point to a nonexistent rrd file (/var/www/cacti/rra/box_infernix_net_traffic_in_88.rrd), for *all* the graphs. Yep, on all traffic input graphs for all hosts, all DEF: lines point to this file.
Is there something that needs to be regenerated or cleared?
Well, I cleared the poller cache a few times, which didn't help. Then I just manually set the Graph Item Inputs for each graph (they were all set to None), but its fortunate that I only have 10 graphs.
I am glad you got things working, but you definitely should not have to "touch" each graph. I addressed a bug for version 0.8.2 that supposedly fixed the exact issue you are having now. I am going to try and re-produce the buggy condition, but you may want to come up with a set of steps to reproduce the problem just in case.
raX wrote:I am glad you got things working, but you definitely should not have to "touch" each graph. I addressed a bug for version 0.8.2 that supposedly fixed the exact issue you are having now. I am going to try and re-produce the buggy condition, but you may want to come up with a set of steps to reproduce the problem just in case.
-Ian
I am running Cacti 0.8.2a and the only way I could fix the error with the DEF's was to delete the Data Source pointing at the template, then when I recreated the DS it worked fine.
I would like to fix this problem, but I cannot seem to find the correct list of steps to reproduce it. Could either of you re-create this problem using a fresh install of 0.8.2a? If so, could you post your steps here?
raX wrote:I would like to fix this problem, but I cannot seem to find the correct list of steps to reproduce it. Could either of you re-create this problem using a fresh install of 0.8.2a? If so, could you post your steps here?
-Ian
Hi Ian,
I haven't had a chance to reinstall cacti from scratch, but I will give it a go tomorrow. Got to build a new monitoring server for the network.
On a different note how fast is the new poller? As we are looking at monitoring approximately 15,000 mib variables every 5 minutes.
Wyvern wrote:On a different note how fast is the new poller? As we are looking at monitoring approximately 15,000 mib variables every 5 minutes.
It is a lot faster than the current php-based poller. Not only is it written in c, but it is multi-threaded as well. Let me know if you decide to go this route because I am more than willing to help you debug any problems that you encounter. I have yet to test cactid with 15,000+ items, but I see no reason why this would not work.