I would try svn, I can't say what's actually in the tar.  Agree with you on 
ports.  Makes me long for the day of static/monolythic exe's. Can't understand 
why some developers insist on using a library with two dozen other dependancies 
for one simple function!  But I digress...

----- Original Message -----
From: [email protected] <[email protected]>
To: [email protected] <[email protected]>
Sent: Thu Apr 29 21:37:21 2010
Subject: Re: [Ntop] No data in Netflow on FreeBSD 7.0

Hmm, well, when would have been within the past two weeks, maybe 
Wednesday last, where - I'm 95% sure from ntop.org. Tar ball, not SVN.

we generally stay away from ports because of the lack of control over 
dependencies. I was going to do the port, but it started pulling 
rrdtool, which I already have from source. I'll figure out how to 
convince the port to use my version of rrdtool tomorrow and give it a try.

On 04/29/2010 10:20 PM, Gary Gatten wrote:
> Where/when did you get your source?  I wanna diff with mine.
>
> ----- Original Message -----
> From: [email protected]<[email protected]>
> To: [email protected]<[email protected]>
> Sent: Thu Apr 29 21:06:00 2010
> Subject: Re: [Ntop] No data in Netflow on FreeBSD 7.0
>
> Recompiled with 3 DEBUG defines uncommented in netflowPlugin.c
>
> Restarting shows the Netflow device being initialized and mapped to an
> ntop device, then after everything else is setup, I see my one packet,
> than nothing else:
>
> Apr 29 21:45:53 netmon ntop[49722]:   NETFLOW_DEBUG: Received NetFlow
> packet(len=1272)(deviceId=1)
> Apr 29 21:45:53 netmon ntop[49722]:   NETFLOW: dissectFlow(len=1272,
> device=1) [flow packet=1]
> Apr 29 21:45:53 netmon ntop[49722]:>>>>  NETFLOW: handleGenericFlow() called
> Apr 29 21:45:53 netmon ntop[49722]:   NETFLOW_DEBUG: a=3232235792
> Apr 29 21:45:53 netmon ntop[49722]:   DEBUG: 192.168.1.16:0 ->
> 192.168.100.23:771 [last=923276096][first=923276096][last-first=0]
> Apr 29 21:46:01 netmon ntop[49722]:   IDLE_PURGE: Device 0 [em0]
> FINISHED selection, 0 [out of 4] hosts selected
> Apr 29 21:46:01 netmon ntop[49722]:   IDLE_PURGE: Device em0: no hosts
> [out of 3] deleted
> Apr 29 21:46:01 netmon ntop[49722]:   IDLE_PURGE: Device 1
> [NetFlow-device.2] FINISHED selection, 0 [out of 4] hosts selected
>
> The sequence of RRD cycles and IDLE_PURGEs then continues, with no more
> packets.
>
> I'm happy to post more logging, but I figure if anyone really wants
> pages of logs, they'll let me know.
>
> And that's enough for me tonight.
>
> tim
>
> On 04/29/2010 09:30 PM, Tim Palmer wrote:
>    
>> I'll give the debug recompile a shot. I'm no coder, but I suspect I
>> can find the debug code.
>>
>> I should have mentioned - I had this same machine working last week
>> with 3.3.10, also under FBSD 7.0, but p9 rather than p12 as now,
>> although if I ran in daemon mode, I got the infamous "kevent: Bad file
>> descriptor" messages. I really wanted daemon mode, so tried the
>> "rfork" change from
>> http://lists.freebsd.org/pipermail/freebsd-hackers/2009-November/030105.html.
>> That gave me the same results I'm getting now - nice screens, no
>> errors, and no data. The box's RAID died a horrible death over the
>> weekend, and I'm working on a rebuild.
>>
>> I hate that feeling when I'm having a problem no one else seems to
>> have. Makes me think I'm missing something simple, but I just can't
>> mail it.
>>
>> thanks again for your time,
>>
>> tim
>>
>> On 04/29/2010 08:22 PM, Gary Gatten wrote:
>>      
>>> I used to run on FBSD 6, but now RHEL5 so don't know first hand of
>>> issues with FBSD anymore. My understanding is it *works* fine.  I
>>> know there is some logic in the code to do somethings different if OS
>>> is FBSD, maybe that is broken?  If you like source, there are several
>>> places in the code to enable DEBUG output, starting in
>>> globals-defines.h, but also within the netflow plugin module code.
>>> It will spew a $HITLOAD of messages, so maybe build it with a
>>> different prefix and only run for a few seconds.  I'll try to think
>>> of something else....  MAYBE try the port just to see if it works and
>>> if so diff the code?
>>>
>>> ----- Original Message -----
>>> From:
>>> [email protected]<[email protected]>
>>> To: [email protected]<[email protected]>
>>> Sent: Thu Apr 29 18:55:34 2010
>>> Subject: Re: [Ntop] No data in Netflow on FreeBSD 7.0
>>>
>>> Thanks for the quick reply(s)
>>>
>>> "Sounds like you tried most everything."
>>> <sigh>   yeah, that's what I thought.
>>>
>>> Interface is selected. Netflow statistics says 1 packet in, 1 out, 40
>>> bytes, 1 flow. Nothing in any of the v1/v5/v9 rows, nothing except "1
>>> flow processed" in the Discarded section.
>>>
>>> running as root, with -t 5 doesn't show me anything I can identify as a
>>> problem - This is the log section that seems most relevant:
>>>
>>> Apr 29 19:40:30 netmon ntop[37164]:   Now running as requested user
>>> 'root' (0:0)
>>> Apr 29 19:40:30 netmon ntop[37164]:   Device  0.
>>> em0                            (active)
>>> Apr 29 19:40:30 netmon ntop[37164]:   Device  1.
>>> NetFlow-device.2               (active)
>>> Apr 29 19:40:30 netmon ntop[37164]:   Note: Reporting device initally
>>> set to 1 [NetFlow-device.2]
>>> Apr 29 19:40:30 netmon ntop[37164]:   MEMORY: Base interface structure
>>> (no hashes loaded) is 0.33MB each
>>> Apr 29 19:40:30 netmon ntop[37164]:   MEMORY:     or 0.65MB for 2
>>> interfaces
>>> Apr 29 19:40:30 netmon ntop[37164]:   MEMORY: ipTraffixMatrix structure
>>> (no TrafficEntry loaded) is 0.36MB
>>> Apr 29 19:40:30 netmon ntop[37164]:   THREADMGMT[t34412171552]: ntop
>>> RUNSTATE: RUN(4)
>>> Apr 29 19:40:30 netmon ntop[37164]:   THREADMGMT[t34412176704]: NPS(1):
>>> Started thread for network packet sniffing [em0]
>>> Apr 29 19:40:30 netmon ntop[37164]:   THREADMGMT[t34412176704]:
>>> NPS(em0): pcapDispatch thread starting [p37164]
>>> Apr 29 19:40:30 netmon ntop[37164]:   THREADMGMT[t34412175968]: NETFLOW:
>>> (port 9990) thread running [p37164]
>>>
>>> and:
>>>
>>> Apr 29 19:40:55 netmon ntop[37164]:   RRD: Cycle 0 ended, 38 RRDs
>>> updated, 0.037 seconds
>>> Apr 29 19:40:55 netmon ntop[37164]:   RRD_DEBUG: Sleeping for 300
>>> seconds (interval 300, end at Thu Apr 29 19:45:55 2010)
>>> Apr 29 19:43:04 netmon ntop[37164]:   SECURITY: Loading items table
>>> Apr 29 19:45:57 netmon ntop[37164]:   RRD: Cycle 1 ended, 18 RRDs
>>> updated, 0.006 seconds
>>>
>>> tim
>>>
>>> On 04/29/2010 06:32 PM, Gary Gatten wrote:
>>>        
>>>> Also, try running as root to rule out perms and maybe start with -t
>>>> 5 and hope to get some useful messages in the log.
>>>>
>>>> -----Original Message-----
>>>> From: [email protected]
>>>> [mailto:[email protected]] On Behalf Of Gary Gatten
>>>> Sent: Thursday, April 29, 2010 4:58 PM
>>>> To: '[email protected]'
>>>> Subject: Re: [Ntop] No data in Netflow on FreeBSD 7.0
>>>>
>>>> Sounds like you tried most everything.
>>>>
>>>> What does "Plugins>    Netflow>    Statistics" show?
>>>>
>>>> Also, have you "Selected" the interface?  "Admin>    Switch NIC" and
>>>> actually choose your netflow interface?
>>>>
>>>> G
>>>>
>>>>
>>>> -----Original Message-----
>>>> From: [email protected]
>>>> [mailto:[email protected]] On Behalf Of Tim Palmer
>>>> Sent: Thursday, April 29, 2010 4:34 PM
>>>> To: [email protected]
>>>> Subject: [Ntop] No data in Netflow on FreeBSD 7.0
>>>>
>>>> Good Day,
>>>>
>>>> I'm trying to get ntop working on a FreeBSD 7.0 amd64 box. I've had
>>>> problems compiling 3.3.10, so tried 3.4pre3.
>>>>
>>>> I'm only interested in seeing data on a NetFlow interface. Nothing
>>>> local
>>>> is needed. However, I'm seeing similar behavior on eth0, only the
>>>> Traffic Statistics table on the Summary Traffic page show very many
>>>> packets dropped by libpcap.
>>>>
>>>> Compile and installation work fine. Ntop starts fine, web interface is
>>>> fully functional. Netflow plugin is enabled and active. But there is
>>>> only one packet shown for the NetFlow device, no packets dropped by
>>>> ntop. I *believe* I've tried all ip address configuration options. Most
>>>> other settings are default. Running in daemon mode does not produce any
>>>> warnings on the console. Listen port is not default, and I've
>>>> configured
>>>> in the web UI, not spec
>>>>
>>>> started with {prefix}/bin/ntop -w 81 -u ntop -L -d
>>>>
>>>> tcpdump shows data coming in on the port I'm expecting it. Ethereal
>>>> confirms they are legit netflow/cflow packets.
>>>>
>>>> sockstat shows ntop listening on the udp4 port expected.
>>>>
>>>> disabling ipfw doesn't help.
>>>>
>>>> Files are created in {prefix}/var/ntop/rrd/interfaces/NetFlow-device.2.
>>>> They are being updated, but only with NANs or 0.000 entries.
>>>>
>>>> Netflow statistics page in the web UI shows just the one packet,
>>>> 56bytes. No dropped flows or other problems.
>>>>
>>>> We prefer to compile from source, so haven't tried the port yet.
>>>>
>>>> rrdtool is 1.4, compiled from source. Cacti is also on this box, and
>>>> has
>>>> no problem w/ rrdtool.
>>>> perl is 5.10.0
>>>>
>>>> flow-capture is also in use on this box (for other devices, on other
>>>> ports) and is working properly.
>>>>
>>>> I'm at a loss. If there's more information I can provide, I am most
>>>> happy to do so.
>>>>
>>>> Kernel is custom. I have not yet tried with GENERIC, but try that next.
>>>>
>>>> FreeBSD xxx.xxx.xxx 7.0-RELEASE-p12 FreeBSD 7.0-RELEASE-p12 #0: Wed Apr
>>>> 28 17:46:20 EDT 2010
>>>> [email protected]:/usr/obj/usr/src/sys/NETMON  amd64
>>>>
>>>> Thank you very much for your time. I'm sort of hoping this is just
>>>> something stupid I'm missing.
>>>>
>>>> Tim Palmer
>>>> _______________________________________________
>>>> Ntop mailing list
>>>> [email protected]
>>>> http://listgateway.unipi.it/mailman/listinfo/ntop
>>>> _______________________________________________
>>>> Ntop mailing list
>>>> [email protected]
>>>> http://listgateway.unipi.it/mailman/listinfo/ntop
>>>> _______________________________________________
>>>> Ntop mailing list
>>>> [email protected]
>>>> http://listgateway.unipi.it/mailman/listinfo/ntop
>>>>
>>>>
>>>>          
>>> _______________________________________________
>>> Ntop mailing list
>>> [email protected]
>>> http://listgateway.unipi.it/mailman/listinfo/ntop
>>> _______________________________________________
>>> Ntop mailing list
>>> [email protected]
>>> http://listgateway.unipi.it/mailman/listinfo/ntop
>>>
>>>        
>> _______________________________________________
>> Ntop mailing list
>> [email protected]
>> http://listgateway.unipi.it/mailman/listinfo/ntop
>>
>>      
> _______________________________________________
> Ntop mailing list
> [email protected]
> http://listgateway.unipi.it/mailman/listinfo/ntop
> _______________________________________________
> Ntop mailing list
> [email protected]
> http://listgateway.unipi.it/mailman/listinfo/ntop
>
>    

_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop

Reply via email to