<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<br>
<br>
Hi Folks!<br>
<br>
First, Steve: thanks for the backup information.<br>
<br>
Second, I went back to the <i>iperf</i> testing -- something
didn't seem quite right but I didn't even know this test existed
until a few days ago, so guess not doing to bad. Ran the test
'backwards' -- switched client and server -- which I hadn't done the
first time. Found there appears to be a discrepancy, though have no
idea what the next step is.<br>
<br>
<br>
<br>
Ran <font size="-1"><font size="-2"><tt>iperf -c 192.168.0.21 -d</tt></font>
</font>
on "BE2", the Backend with which I have been issues with long
pausing and sluggish FF, playback appears fine.<br>
<br>
This is the results from "M58", which is not used as a Frontend:<br>
<blockquote><font size="-2"><tt>barry@Lenovo-M58:~$ iperf -s</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
<font size="-2"><tt>Server listening on TCP port 5001</tt></font><br>
<font size="-2"><tt>TCP window size: 85.3 KByte (default)</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
<font size="-2"><tt>[ 4] local 192.168.0.21 port 5001 connected
with 192.168.0.4 port 57116</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
<font size="-2"><tt>Client connecting to 192.168.0.4, TCP port
5001</tt></font><br>
<font size="-2"><tt>TCP window size: 162 KByte (default)</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
<font size="-2"><tt>[ 6] local 192.168.0.21 port 33356 connected
with 192.168.0.4 port 5001</tt></font><br>
<font size="-2"><tt>[ ID] Interval Transfer Bandwidth</tt></font><br>
<font size="-2"><tt>[ 6] 0.0-10.1 sec 108 MBytes 89.6
Mbits/sec</tt></font><br>
<font size="-2"><tt>[ 4] 0.0-10.6 sec 35.4 MBytes 28.1
Mbits/sec</tt></font><br>
<font size="-2"><tt>[ 5] local 192.168.0.21 port 5001 connected
with 192.168.0.13 port 50268</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
</blockquote>
<font size="-2">One direction the data travels about 3x faster than
the other -- is this resulting in the sluggishness on the FF
request?<br>
<br>
<br>
</font><font size="-2"><tt><br>
</tt></font><font size="-2">Ran the same test on "BE1", the old
Backend which has the proper and snappy Fast Forwarding; results;
results as viewed on "M58" (same server as above test) :</font><font
size="-2"><tt><br>
</tt></font>
<blockquote><font size="-2"><tt>Client connecting to 192.168.0.13,
TCP port 5001</tt></font><br>
<font size="-2"><tt>TCP window size: 153 KByte (default)</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
<font size="-2"><tt>[ 6] local 192.168.0.21 port 51130 connected
with 192.168.0.13 port 5001</tt></font><br>
<font size="-2"><tt>[ 6] 0.0-10.0 sec 112 MBytes 93.4
Mbits/sec</tt></font><br>
<font size="-2"><tt>[ 5] 0.0-10.0 sec 106 MBytes 88.5
Mbits/sec</tt></font><br>
<font size="-2"><tt>[ 4] local 192.168.0.21 port 5001 connected
with 192.168.0.13 port 50270</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
<font size="-2"><tt>Client connecting to 192.168.0.13, TCP port
5001</tt></font><br>
<font size="-2"><tt>TCP window size: 153 KByte (default)</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
<font size="-2"><tt>[ 6] local 192.168.0.21 port 51132 connected
with 192.168.0.13 port 5001</tt></font><br>
<font size="-2"><tt>[ 4] 0.0-10.0 sec 106 MBytes 89.0
Mbits/sec</tt></font><br>
<font size="-2"><tt>[ 6] 0.0-10.0 sec 112 MBytes 93.5
Mbits/sec</tt></font><br>
<font size="-2"><tt>[ 5] local 192.168.0.21 port 5001 connected
with 192.168.0.4 port 57204</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
</blockquote>
<font size="-2">So on the old/original Backend (BE1) the test
results are almost the same in both directions. <br>
<br>
<br>
Ran the test on BE2 again:</font><font size="-2"><tt><br>
</tt></font>
<blockquote><font size="-2"><tt>Client connecting to 192.168.0.4,
TCP port 5001</tt></font><br>
<font size="-2"><tt>TCP window size: 170 KByte (default)</tt></font><br>
<font size="-2"><tt>------------------------------------------------------------</tt></font><br>
<font size="-2"><tt>[ 6] local 192.168.0.21 port 33702 connected
with 192.168.0.4 port 5001</tt></font><br>
<font size="-2"><tt>[ 6] 0.0-10.1 sec 110 MBytes 90.7
Mbits/sec</tt></font><br>
<font size="-2"><tt>[ 5] 0.0-10.6 sec 39.4 MBytes 31.1
Mbits/sec</tt></font><br>
</blockquote>
A hare faster for some reason but still significantly lopsided.<br>
<br>
(Might also explain why viewing on the BE [BE2] I'm not seeing the
FF issue ==> 'internal connection'.)<br>
<br>
So thanks in advance! Probably made the presentation more
complicated than necessary but I have next to no idea what I'm
doing.<br>
<br>
Barry<br>
<br>
<br>
P.S.:<br>
<br>
192.168.0.4 = BE2 (Backend 2), the 'side' with the sluggish
Fast Forwarding.<br>
192.168.0.13 = BE1 (Backend 1), the 'side' that's old and
works fine but will need to be replaced because of age.<br>
The two Backends are independent. Frontends connected to BE1
work fine, same FE to BE2 exhibit the FF sluggishness.<br>
192.168.0.21 = M58. Is <u>not</u> a FE; is my desk
computer.<br>
<br>
</body>
</html>