<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Recent changes to support-requests</title><link>https://sourceforge.net/p/fprobe/support-requests/</link><description>Recent changes to support-requests</description><atom:link href="https://sourceforge.net/p/fprobe/support-requests/feed.rss" rel="self"/><language>en</language><lastBuildDate>Mon, 28 Feb 2005 18:21:54 -0000</lastBuildDate><atom:link href="https://sourceforge.net/p/fprobe/support-requests/feed.rss" rel="self" type="application/rss+xml"/><item><title>Memory use?</title><link>https://sourceforge.net/p/fprobe/support-requests/9/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;Is this normal?&lt;/p&gt;
&lt;p&gt;5589 _fprobe    4    0  355M  335M sleep    bpf  &lt;br /&gt;
147:43  2.00% fprobe&lt;/p&gt;
&lt;p&gt;Seems quite high for only tracking 5mbps.&lt;/p&gt;
&lt;p&gt;CMD line is:&lt;/p&gt;
&lt;p&gt;/usr/local/sbin/fprobe -f ip -i em1 -n 5 -q 70000 -a &lt;br /&gt;
XXX.XXX.XXX.XXX -b 32000 -c /var/empty -u _fprobe -l 1 &lt;br /&gt;
XXX.XXX.XXX.XXX:XXXX&lt;/p&gt;
&lt;p&gt;Please email response to: admin@wolfpaw.com&lt;/p&gt;
&lt;p&gt;Thanks!&lt;br /&gt;
D,&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Anonymous</dc:creator><pubDate>Mon, 28 Feb 2005 18:21:54 -0000</pubDate><guid isPermaLink="false">https://sourceforge.netd5125192c97efe30d190fb6c457e20a541222ecc</guid></item><item><title>Is netflow timefield msecs or secs ?</title><link>https://sourceforge.net/p/fprobe/support-requests/8/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;I was wondering what time field value fprobe-ulog is&lt;br /&gt;
using for NETFLOW.&lt;/p&gt;
&lt;p&gt;In nfacctd I can set the config file to this:&lt;/p&gt;
&lt;p&gt;KEY:        nfacctd_time_secs&lt;br /&gt;
VALUES:     [true|false]&lt;br /&gt;
DESC:       makes 'nfacctd' expect times included in Netflow&lt;br /&gt;
header to be in seconds instead of msecs. This&lt;br /&gt;
seems to be a quite common case. (default: false;&lt;br /&gt;
times are expected in msecs)&lt;/p&gt;
&lt;p&gt;KEY:        nfacctd_time_new&lt;br /&gt;
VALUES:     [true|false]&lt;br /&gt;
DESC:       makes 'nfacctd' to ignore times included in&lt;br /&gt;
Netflow header and makes it build new times. Although&lt;br /&gt;
this could be useful to avoid date skews, remember&lt;br /&gt;
that newly generated times will be strictly connected&lt;br /&gt;
with the arrival of Netflow packets. (default: false)&lt;/p&gt;
&lt;p&gt;nfacctd_time_new I try not to use, I will start&lt;br /&gt;
ntpclient on WRT54G (OpenWRT), so the date is&lt;br /&gt;
up-to-date all the time.&lt;br /&gt;
But for nfacctd_time_secs I'm wondering.&lt;/p&gt;
&lt;p&gt;Cheers&lt;/p&gt;
&lt;p&gt;Blackvel&lt;br /&gt;
E-Mail&lt;br /&gt;
Thomas Heiss&lt;br /&gt;
contact &amp;lt;&amp;lt;at&amp;gt;&amp;gt; agency ||dot|| franken &amp;lt;dot&amp;gt; de&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Anonymous</dc:creator><pubDate>Sat, 18 Sep 2004 17:02:56 -0000</pubDate><guid isPermaLink="false">https://sourceforge.netbfc4d580510b8634acc908620c9b99edb5fd809c</guid></item><item><title>fprobe-ulog not running on WRT54G, code needs to be cleaned</title><link>https://sourceforge.net/p/fprobe/support-requests/7/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;Hello,&lt;/p&gt;
&lt;p&gt;I tried to run fprobe on Linksys WRT54G with Linux &lt;br /&gt;
OpenWRT on it (cross-compile for mipsel).&lt;/p&gt;
&lt;p&gt;fprobe is working, but has the problem with libpcap that &lt;br /&gt;
it get's the wrong src_ip and dst_ip (not the internal IP, &lt;br /&gt;
but the already masqueraded IP from NAT router).&lt;/p&gt;
&lt;p&gt;Also I am not sure, if it supports setting netflow input &lt;br /&gt;
and output SNMP indexes.&lt;br /&gt;
I try to use nfacctd (pmacctd) to collect netflow ip &lt;br /&gt;
accounting data on #2 linux server.&lt;br /&gt;
And of course, I want to set the agent_id depending in &lt;br /&gt;
what direction the packets are flowing (input or either &lt;br /&gt;
output).&lt;/p&gt;
&lt;p&gt;So I decided to test fprobe-ulog, because with ULOG, it &lt;br /&gt;
should get the right src_ip and dst_ip (the internal, not &lt;br /&gt;
the masqueraded).&lt;/p&gt;
&lt;p&gt;However I can't start fprobe-ulog on WRT, it fails with &lt;br /&gt;
message:&lt;br /&gt;
malloc(): unable to allocate memory&lt;/p&gt;
&lt;p&gt;After taking a look into the fprobe-ulog.c source code, &lt;br /&gt;
this error message can be thrown by various code &lt;br /&gt;
places since it uses the errmalloc: goto in the whole &lt;br /&gt;
source file.&lt;/p&gt;
&lt;p&gt;Can you please cleanup the source code in the next &lt;br /&gt;
versions, so a more detailed message can be displayed &lt;br /&gt;
and I also know, what is going wrong really ?&lt;/p&gt;
&lt;p&gt;I can't find the place, which causes the malloc() &lt;br /&gt;
problem :)&lt;/p&gt;
&lt;p&gt;Thank you.&lt;/p&gt;
&lt;p&gt;Thomas Heiss&lt;br /&gt;
contact &amp;lt;at| agency &amp;lt;&amp;lt;dot&amp;gt;&amp;gt; franken &amp;lt;&amp;lt;dot&amp;gt; de&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Anonymous</dc:creator><pubDate>Mon, 13 Sep 2004 10:54:57 -0000</pubDate><guid isPermaLink="false">https://sourceforge.nete580bf5cdf42d22d0868fda202508ceef79eadd0</guid></item><item><title>cannot use with tun device</title><link>https://sourceforge.net/p/fprobe/support-requests/6/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;can it work with tun device?&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Anonymous</dc:creator><pubDate>Sat, 08 May 2004 08:11:01 -0000</pubDate><guid isPermaLink="false">https://sourceforge.netec285885232d0c45767239ea70fa2d1ba228832a</guid></item><item><title>Autofilter by MAC address in multiaccess interfaces</title><link>https://sourceforge.net/p/fprobe/support-requests/5/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;While testing this software I receive flows thit not&lt;br /&gt;
olny incomning (from net to PC/gateway) packets but&lt;br /&gt;
also outbound packets. This doubles counted traffic.&lt;br /&gt;
As I understand Cisco's NetFlow, cisco routers count&lt;br /&gt;
only incoming traffic.&lt;br /&gt;
There is woraround - add "ether src MACADDR" to filter&lt;br /&gt;
rules in command line.&lt;/p&gt;
&lt;p&gt;Think, is isn't reliable enough. Command line switch&lt;br /&gt;
"count only incoming/ounbound/both" is reasonable.&lt;br /&gt;
("both" is usefull to automate filtering out odd&lt;br /&gt;
traffic on promic. interface.&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Victor Sheldeshov</dc:creator><pubDate>Fri, 06 Feb 2004 11:13:19 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net670138a04e7e5d7b972e60cc85fcf3354226c009</guid></item><item><title>An idea for SNMP If number on Trunk interface</title><link>https://sourceforge.net/p/fprobe/support-requests/4/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;If we listen for a .1q Trunk interface it is reasonable&lt;br /&gt;
to unpack VlanID tag.&lt;br /&gt;
Using VlanId we can craft a SNMP If number in one or&lt;br /&gt;
other manner (adding to a provided in command line&lt;br /&gt;
number or concatination with it).&lt;br /&gt;
... If we can unpack VlanID tag...&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Victor Sheldeshov</dc:creator><pubDate>Wed, 04 Feb 2004 14:08:18 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net00433662e425eea8ab39631d08e5f38e770ce4ea</guid></item><item><title>interface specification</title><link>https://sourceforge.net/p/fprobe/support-requests/3/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;Hello, I was using other tools to display summaries of the &lt;br /&gt;
flow information from fprobe and it seems that they are &lt;br /&gt;
getting confused on the input and output index number. &lt;br /&gt;
Fprobe seems to label them both as interface 0 which is &lt;br /&gt;
probably valid as a host isn't a router. Is there a work &lt;br /&gt;
around for this so that the incoming and outgoing &lt;br /&gt;
packets can be differentiated? thanks&lt;/p&gt;
&lt;p&gt;john@galleries.ab.ca&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Anonymous</dc:creator><pubDate>Fri, 07 Nov 2003 09:11:30 -0000</pubDate><guid isPermaLink="false">https://sourceforge.netc3bbccee48a82fb2254bceaf9b21e89abb49406f</guid></item><item><title>fprobe daemon response</title><link>https://sourceforge.net/p/fprobe/support-requests/2/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;I am running redhat version 6.2 and are loading the &lt;br /&gt;
daemon with the following command, ./fprobe -ieth0 -l2 -&lt;br /&gt;
fip localhost:2055 which seems to launch the daemon &lt;br /&gt;
successfully in debug mode. The problem I am having is I &lt;br /&gt;
always get zero flow counts, nothing is reporting on the &lt;br /&gt;
screen either. Any advice is appreciated. thanks &lt;br /&gt;
john@galleries.ab.ca&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Anonymous</dc:creator><pubDate>Wed, 15 Oct 2003 03:19:42 -0000</pubDate><guid isPermaLink="false">https://sourceforge.net991719e7a483b9f54ed003093900ce3de7ec9e92</guid></item><item><title>weird issues !!!!!!!!!!!!!!!!!!!!!!!!!!!!!</title><link>https://sourceforge.net/p/fprobe/support-requests/1/</link><description>&lt;div class="markdown_content"&gt;&lt;p&gt;runnning fprobe v1.0 Pre4 on Redhat 7.3 with 3com &lt;br /&gt;
Fiber Gig NIC.  Also installed is cflowd 2.0.&lt;/p&gt;
&lt;p&gt;I monitor Gig NIC (eth1) using tcpdump and see various &lt;br /&gt;
flows including unicast packets (telnet, etc..), and of &lt;br /&gt;
course broadcast packets.  Fprobe is does not appear &lt;br /&gt;
to be exporting all the flows that I am viewing using &lt;br /&gt;
tcpdump, especially unicast flows.  The gig interface is &lt;br /&gt;
running at around 40 Mbps.  We are not forwarding any &lt;br /&gt;
of the frames on the gig interface.&lt;/p&gt;
&lt;p&gt;Please HELP!, with any insight and feedback!!&lt;/p&gt;
&lt;p&gt;eric1462@yahoo.com&lt;/p&gt;&lt;/div&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">eric sandoval</dc:creator><pubDate>Fri, 09 May 2003 01:04:45 -0000</pubDate><guid isPermaLink="false">https://sourceforge.netd7426e9396485140d2212fa942a217a863751920</guid></item></channel></rss>