Cacti Graph broken many times.
 

Hi,
I have been using Cacti 0.8.7e but having some problem that it was a little old server and many templates and extensions are not working properly. Moreover it is a 32 bit system.
Then we decided to install a new system and new cacti 0.8.7i. All going well but some of our server's having high load causing problem. They appears different graphs other than our old one.
Below are the screenshots and please check them.

new cacti -->http://db.tt/CpaRfpbz
old cacti-->http://db.tt/W221nZp9
To get rid of this problem I have checked all the permission files,install the cacti twice but still the same situation. Also noted that my previous cacti is installed on Centos 5.4 32 bit and new cacti on Debian 6 64 bit system.
Also noted that interface statistics, CPU Utilization or Memory Usage all appear the same/exact but only the Disk Space(Mounted Partitions) causing problems.
Your Kind response are highly appreciated.
Thanks in advance.

So what exactly is your problem?
It looks like you've runn both Nagios servers at the same time and compared the graphs.
They're not identical, but that's not an issue.
You will get the odd drop out (white space on graph) when the network cxn to that sever is heavily loaded.
The gaps you've got are very small and not worth worrying about.
The main one appears to be about 05.30, so what are the systems/network doing at that at time.
A wild guess says backups; very i/o bound...

Hey Chris,
Thanks for your response.
So my problem exactly is that the difference between my two servers which make people confuse.
And I can assure you on that time no network outage occurs or server rebooted. My all other graphs are appears well.
I've stopped Nagios on that new cacti server and observe last day. Here are screenshots for you.
Here see broken graphs in New Cacti server.
cacti new -1 -->http://db.tt/AibshTPO
This view is quite confusing,memory buffer looks clear but real memory seems broken.
cacti new -2 -->http://db.tt/x3o5ZRpO
Here only one broken image--no server rebootig/nerwork failure issue there.
cacti old -1 -->http://db.tt/SluFNEID
here you see more of them.
cacti old -2 -->http://db.tt/8OrlRyqT
And more.
cacti old -3 -->http://db.tt/ndnglgs1

I've not understand what you mean. Can you please explain a bit more?
--------------
Thanks
Ahsan

That's not what I said. I said a heavy network or other I/O load.
Backups for example are notorious for this and often happen very early in the morning, as this is usually the time the systems are least used, especially if they are used by humans.

if you find the gaps appear at more or less the same time every eg day, then you should find out what else is happening at that time.
However(!), a short gap as you seem to have is nothing to worry about.

Feel free to check with the Cacti forums :)
http://forums.cacti.net/

BTW, there's not much point in comparing servers as you are; they will not have identical results unless you have a very small & lightly loaded LAN.