Hi, i've been experimenting with the diskfree.pl script, but i can not get it working.. i`m trying to make a graph like this. Is there a tutorial for graphs like these?
Regards
Jan Peter 'Banaan' Balkenende
Disk Space Graph
Moderators: Developers, Moderators
If you run "perl diskspace.pl /" (or /usr etc) from a shell does it show the proper values?
I collect disk space information on my system as well, but I show individual usage for web, mail and database on the file system
http://status.digitalorphans.org/graphs.php/disk
There's two pieces to this: I run cmd.php as a non-root user, which doesn't have access to all of the directories. So, I have a root cronjob setup to get the sizes, using du:
Each file just simple contains a number, which is the number of bytes.
Then, in cacti, i have a data input called "get file contents", using:
And for each data source, I'll specify "file" as "/var/stats/size_db" etc.
If you run cmd.php as root, then you could just make the data input string:
The difference with my script though, is it shows bytes used, not bytes free, but it's handy because it shows exactly how much space everything is using.
I collect disk space information on my system as well, but I show individual usage for web, mail and database on the file system
http://status.digitalorphans.org/graphs.php/disk
There's two pieces to this: I run cmd.php as a non-root user, which doesn't have access to all of the directories. So, I have a root cronjob setup to get the sizes, using du:
Code: Select all
*/5 * * * * du -sb /var/lib/mysql | awk '{print $1 }' > /var/stats/size_db
*/5 * * * * du -sb /home/httpd/vhosts | awk '{print $1 }' > /var/stats/size_web
*/5 * * * * du -sb /var/qmail/mailnames | awk '{print $1 }' > /var/stats/size_mail
Then, in cacti, i have a data input called "get file contents", using:
Code: Select all
Input string: cat <file>
Output string: <data>
If you run cmd.php as root, then you could just make the data input string:
Code: Select all
du -sb <file> | awk '{print $1 }'
Hi GregM,
Your graph looks fabulous. I would really want to setup the same on my SME linux box (www.e-smith.org).
I have installed Cacti and find it immensely useful. I need more graphs than are standard though.
How do I collect the data-input 'how much usage each user's html-folder structure is', and showing it in Cacti?
I have an SME-Server 5.6 (Linux) downloaded from www.e-smith.org and set-up cacti from this How-to: http://www.speedfactor.ath.cx/howto/
And have tried to setup additional graphs according to this How-to: http://www.speedfactor.ath.cx/howto/snmp.html
On the SME-Server you divide the different websites hosted up into Information-bays. I would like to view graphically how much each I-bay takes up. Is there a way to do this in Cacti?
BR
Morten
Your graph looks fabulous. I would really want to setup the same on my SME linux box (www.e-smith.org).
I have installed Cacti and find it immensely useful. I need more graphs than are standard though.
How do I collect the data-input 'how much usage each user's html-folder structure is', and showing it in Cacti?
I have an SME-Server 5.6 (Linux) downloaded from www.e-smith.org and set-up cacti from this How-to: http://www.speedfactor.ath.cx/howto/
And have tried to setup additional graphs according to this How-to: http://www.speedfactor.ath.cx/howto/snmp.html
On the SME-Server you divide the different websites hosted up into Information-bays. I would like to view graphically how much each I-bay takes up. Is there a way to do this in Cacti?
BR
Morten
It really depends on the method your server uses to store everything for users. Just check the disk space for that directory:
Set it up as a shell script that accepts a parameter for username, and then just make a data input for each user using the parameter to select which user.
Alternatively, you could probably automate it a bit by going around cacti, and just use rrdtool directly, using your actual list of users to generate everything, but that could be a lot more work depending if you can program or not.
Code: Select all
du -sb /home/vhosts/username | awk '{print $1 }'
Alternatively, you could probably automate it a bit by going around cacti, and just use rrdtool directly, using your actual list of users to generate everything, but that could be a lot more work depending if you can program or not.
diskfree.pl issue
i have a lvm array on /dev/anthl_vg/anthl_lv.
# df
S.ficheros Bloques de 1K Usado Dispon Uso% Montado en
/dev/hda1 9646700 8608512 1038188 90% /
/dev/anthl_vg/anthl_lv
234423304 119496092 114927212 51% /mnt/raid
the carriage return is caused by default on df. You should run df -P (stands for Posix'ly Correct) so output is on the same line an the diskfree.pl script parses it correctly.
# df -P
S.ficheros Bloques de 1024 Usado Dispon Ocupado Montado en
/dev/hda1 9646700 8608528 1038172 90% /
/dev/anthl_vg/anthl_lv 234423304 119496548 114926756 51% /mnt/raid
# df
S.ficheros Bloques de 1K Usado Dispon Uso% Montado en
/dev/hda1 9646700 8608512 1038188 90% /
/dev/anthl_vg/anthl_lv
234423304 119496092 114927212 51% /mnt/raid
the carriage return is caused by default on df. You should run df -P (stands for Posix'ly Correct) so output is on the same line an the diskfree.pl script parses it correctly.
# df -P
S.ficheros Bloques de 1024 Usado Dispon Ocupado Montado en
/dev/hda1 9646700 8608528 1038172 90% /
/dev/anthl_vg/anthl_lv 234423304 119496548 114926756 51% /mnt/raid
Who is online
Users browsing this forum: No registered users and 4 guests