HTTP tunneling issue

28 views
Skip to first unread message

Sam F

unread,
Oct 20, 2010, 2:51:36 AM10/20/10
to BigBlueButton-dev
Hello,

I don't want to spam the bigbluebutton team and every user but, I have
a problem using the HTTP tunneling feature on a Linux Client.

I have already opened an issue (668) about that, however here is some
details in order to speak about and see if any body had this problem
too.


1 - So, I was trying the Http tunneling with the 0.70 version
installed on my own server.
I set an iptables rule in order to block the 1935 RTMP port.
I joined the demo conference and after 15s, I was logged out and got
these logs from the client :

10/15/2010 18:14:52.728 [DEBUG] NetConnectionDelegate:Connection to
viewers application closed
10/15/2010 18:14:52.728 [DEBUG] [ViewersSOService]:Disconnected from
the Viewers application
10/15/2010 18:14:52.728 [DEBUG] Sending Viewers REMOVE_WINDOW message
to main
10/15/2010 18:14:52.728 [DEBUG] routing message to TO_MAIN_APP
10/15/2010 18:14:52.729 [DEBUG] MainEndpointMediator::Got
REMOVE_WINDOW from FROM_VIEWERS_MODULE
10/15/2010 18:14:52.751 [DEBUG] routing message to TO_MAIN_APP
10/15/2010 18:14:52.752 [DEBUG] MainEndpointMediator::Got
USER_LOGGED_OUT from FROM_VIEWERS_MODULE


Nothing relevant from the server's logs.


2 - I have been looking for a solution because I'm only working with
Linux Client...
I tested the 0.71 dev version and the pb is still here.

More over, I found these information in /var/log/nginx/error.log,
hoping it will help somebody to understand the pb.

error.log
==========
2010/10/19 09:25:23 [error] 8968#0: *7174 recv() failed (104:
Connection reset by peer) while reading upstream, client: 10.1.8.161,
server: WebconfTest.dhcp.lvl.intranet, request: "POST /idle/
11599267/91 HTTP/1.1", upstream: "http://127.0.0.1:8088/idle/
11599267/91 host: "webconftest.dhcp.lvl.intranet"

2010/10/19 09:25:38 [error] 8968#0: *6890 recv() failed (104:
Connection reset by peer) while reading upstream, client: 10.1.8.161,
server: WebconfTest.dhcp.lvl.intranet, request: "POST /idle/
24878199/127 HTTP/1.1", upstream: "http://127.0.0.1:8088/idle/
24878199/127 host: "webconftest.dhcp.lvl.intranet"

2010/10/19 09:25:39 [error] 8968#0: *7294 recv() failed (104:
Connection reset by peer) while reading upstream, client: 10.1.8.161,
server: WebconfTest.dhcp.lvl.intranet, request: "POST /idle/
5504160/129 HTTP/1.1", upstream: "http://127.0.0.1:8088/idle/
5504160/129 host: "webconftest.dhcp.lvl.intranet"

2010/10/19 09:25:40 [error] 8968#0: *7294 recv() failed (104:
Connection reset by peer) while reading upstream, client: 10.1.8.161,
server: WebconfTest.dhcp.lvl.intranet, request: "POST /idle/
25661040/162 HTTP/1.1", upstream: "http://127.0.0.1:8088/idle/
25661040/162 host: "webconftest.dhcp.lvl.intranet"

for each error, the server response code is: "400 5"

10.1.8.161 - - [19/Oct/2010:09:25:23 +0200] "POST /idle/11599267/91
HTTP/1.1" 400 5 "-" "Shockwave Flash"
10.1.8.161 - - [19/Oct/2010:09:25:23 +0200] "POST /close/11599267/92
HTTP/1.1" 200 1 "-" "Shockwave Flash"

And the client close the session right after

0.1.8.161 - - [19/Oct/2010:09:25:38 +0200] "POST /idle/24878199/127
HTTP/1.1" 400 5 "-" "Shockwave Flash"
10.1.8.161 - - [19/Oct/2010:09:25:38 +0200] "POST /close/24878199/128
HTTP/1.1" 200 1 "-" "Shockwave Flash"

10.1.8.161 - - [19/Oct/2010:09:25:39 +0200] "POST /idle/5504160/129
HTTP/1.1" 400 5 "-" "Shockwave Flash"
10.1.8.161 - - [19/Oct/2010:09:25:39 +0200] "POST /close/5504160/130
HTTP/1.1" 200 1 "-" "Shockwave Flash"


3 - I noticed that happen only on a Linux Client, no matter the
distrib, the JDK or the flash plugin

Indeed, I have run some tests from Windows XP, 7, Ubuntu (x64/x86)
with Sun or OpenJDK, Flash 32 or 64 from sevenmachines pppa or
nsdriver, and even from Fedora 13_x64 with Flash from adobe Labs ....)



Does any body have a clue about that ?

Thx a lot !

Fred Dixon

unread,
Oct 20, 2010, 5:55:46 AM10/20/10
to bigblueb...@googlegroups.com
Hi Sam,

You are spamming (though your heart is in the right place).

Seriously, we've got this issue tracked at

http://code.google.com/p/bigbluebutton/issues/detail?id=651

Making the developers bounce back and forth between the issue tracking
system and the bigbluebutton-dev is not helping us fix the problem.

I've copied your information below into issue #651. Please post
additional updates to that issue. We've flagged this as an issue for
0.71.

Regards,... Fred

> --
> You received this message because you are subscribed to the Google Groups "BigBlueButton-dev" group.
> To post to this group, send email to bigblueb...@googlegroups.com.
> To unsubscribe from this group, send email to bigbluebutton-...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/bigbluebutton-dev?hl=en.
>
>

Reply all
Reply to author
Forward
0 new messages