Hi Burton,

Thanks for your response. I hadn't yet looked in the code, as I am not much of a coder and had only been using the pre-compiled RPM packages so far. The problem seems to be affecting every column, which is why I hadn't mentioned a specific one - certainly those entitled bytesSent, bytesRcvd, ipBytesSent and ipBytesRcvd are affected, and most probably the others as well. This seemed to me to be too glaring a problem to have gone unnoticed - which is what lead me to believe that it might have been a simple misconfiguration on my part, rather than a bug in the code itself.

Anyway, I will have a look at the code to see if it's something that can be obviously fixed - but if you have any other thoughts, I'd be most grateful to hear of them.

Best regards,
Giancarlo


On 13/08/2006, at 10:52 PM, Burton Strauss wrote:

I think you've nailed it - now look in the code. 
 
There are two output formatting routines, one for integer (32bit) items and one for 'value' (64bit) items.  Probably the output for the specific one(s) you are seeing the problem in are wrong.  Typically in the buggy ones you will see the 32bit called with the xxx.value field.
 
Since you didn't bother to tell us which column(s) are wrong, I'm not going to guess.
 
Use the tag text (column name) to find the generation routine (i.e. it's the 23rd item).  When you find it, report the specific patch on ntop-dev (or send the patch itself to Luca).
 
-----Burton


From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] On Behalf Of Giancarlo Salamanca
Sent: Sunday, August 13, 2006 4:49 AM
To: [email protected]
Subject: [Ntop] Overflow Issues with dumpData.html

Hi all,

Apologies if this is an issue that has already been covered - I had a look around to find any references to it, but my searches came up short. So, here's my dilemma!

After recently having a play around with ntop running on CentOS 4.2 - currently using version 3.2 from the Dag Apt RPM Repository [i686-redhat-linux-gnu], I've noticed something weird happening with the total bytes tallies as reported in dumpData.html - when looking at hosts with less than a few GB of data, dumpData shows the correct values - but for hosts with more than a few GB, eg. tens or hundreds of GB, the values in dumpData are way off - always showing less than 3 or 4 GB in total traffic. This seems to me very much like a 32-bit integer overflow issue, but the weird thing is the data usage is reported correctly in the web interface of ntop itself, it's just in dumpData that the value is wrong.

So, my question is - is there any way to get around this problem? Hopefully it's a simple fix, and others have experienced it also - so I thought I'd get the opinions of the experts on here :-)

Thanks to all for any advice -
Giancarlo
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop

Reply via email to