Greetings!

We have ntop v.3.3.5 running on a server. This server has had ntop
running on it for quite a while (an earlier version, not sure which, 2.x
most likely) when all of a sudden it started dying on us with no reason.
I thought that it might be because I had set up a cron job to run tshark
(wireshark) on the same interfaces to do 5-minute captures on the
quarter-hours, but even disabling the cron had no effect. I also
upgraded the ntop to the version it's at now, but it still dies for no
apparent reason.

I set it up to do a level 5 debug trace, but could not see anything that
was immediatly apparent from the stdout. Running in daemon mode had no
effect, it dies after anywhere from 30 seconds to 20 minutes after being
started.

The only questionable error on stdout is:
 
  **ERROR** Sanity check failed (2) [Low memory?]

but memory at the time that message was spit out was still about 300Mb
with nothing from swap even touched. When it dies, it frees up about
30Mb of memory.

I also saw a few of these, but they never seem to coincide with the
crash:
 
  *ERROR** NULL host detected

Things I tried that did *not* fix the problem:
 -x, I set it to 4096
 -M, multiple vs. single interfaces
 -s, no-promiscuous
 -u, (ran as root and "nobody")
 -c, host tracking

The box is a Fedora Core 7, 2/6 kernel. 1Gb memory with 2Gb swap. Tons
of disk space.

Any hints? Beuller?

Thanks, as always.

....k 
-=-=-=- 
Kevin Freels
Director of Information Technology
Sendmail, Inc.
[EMAIL PROTECTED]  510/594.5572
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop

Reply via email to