The way it seems to work is the first time it runs you can access on http: port 3000
or https: port 3001 on any address so you can configure. After that initial run it will
only open ports as per /etc/ntop.conf. You missed that opportunity as a result of the
start/stop problems. At least that's the way it appeared to me.
However all is not lost. Just edit /etc/ntop.conf manually. Mine is below. As you can
see I only allow https: access on one address and only local for http:
John, Could be my mistake and not yours
I was assuming he wanted to run the actual ethtool command from my reading of his append as I
stated in my response. I may have got the wrong impression. However, if he just wants to imbed
the ethtool parameters, then yours is the more elegant solution.
As is often the case with Linux - there is more than one solution...
Just put your ethtool command in "/etc/rc.d/rc.local" to run at startup only
Put in "/etc/clearos/firewall.d/local" to run every time the firewall restarts
(which I would think preferred)
"/etc/sysconfig/network-scripts/ifcfg-eth0" is for interface parameters - not
for running a script or comand such as ethool with options
OK - the problem is with /etc/rc.d/init.d/ntop
It's looking for the pidfile in the wrong location and not finding it thinks it is
not running and hence no need to stop...
Just edit /etc/rc.d/init.d/ntop to look like this extract
This was checked using the i686 version - will check if the 64-bit version has the same problem...
OK - installed ntop and used the instructions at "https://www.howtoforge.com/installing-and-configuring-ntop"
All went OK - except for one problem which seems to relate to what you experienced...
# service ntop stop
# service ntop restart
do not work - I had to use
# pkill ntop
# ntop start
This would seem to explain your problem of an existing process when trying to start another.
Maybe a problem with /etc/rc.d/init.d/ntop or /usr/sbin/ntop ?
Drew, It is a few years since I installed ntop - so am not familiar with any present problems
and no longer use it, though if I have time will give it a go...
However, another program you might like to consider is vnStat and a php frontend that will
integrate with the web-server.
Instructions are at "http://danda.poweredbyclear.com/master-frame-vnstat.html"
Examples of output can be found by clinking on "Click here" for the systems in the left-hand
frame and then clicking on "vnStat with vnStat PHP frontend"
By having none ClearOS repositories enabled by default you are risking "rpm hell". Big no no
Immediately go to your /etc/yum.repos.d/ directory and ensure that every None ClearOS repository
(.repo file) has the the line "enabled = 0" for each repo.
That should get rid of the error message - you are pointing to a repository not available which caused
the command to bomb. The rpms you need are in the standard ClearOS 5.2 repository and the command
I gave works for a correctly setup ClearOS 5.2 install...
You need to understand why the files dis-appeared - but you might be able to brute force it back working by :-
# rpm -e clearsdn-intrusion-protection snort app-snort app-snortsam
# yum install clearsdn-intrusion-protection snort app-snort app-snortsam
# chkconfig snort on
# chkconfig snortsam on
# service snort start
# service snortsam start
then to check
# service snort status
# service snortsam status
@Michael - forums definitely loading faster... re-ran same tests...
Initial load now about 4 seconds faster 15 -> 11
Reload about 5.5 seconds faster 12.5 -> 7
Whirlpool about the same as last time
Still need almost a 5x improvement to match Whirlpool