Red5 + CentOS

sh4ka · 05-04-2011, 09:13 AM

Hi everybody,

The problem I'm facing is web scripts that need to connect to the server can't find a way to connect to port 1935, it seems to be blocked in some how...

Also, testing from the red5 test page (http://server.ip.address:5080/demos/port_tester.html) that tests the red5 common ports is also failing, when it should success.

Please note that:

* There is no firewall running on the server.
* Selinux has been disabled.
* No tcpwrappers rules working.
* Red5 service is up and running.

As said before, service is running and connections are responding, from inside and outside the server. But when I connect from any web scripts, it seems to be hang there and the connection is never stablished.

Code:

[root@my.server.com:~]service red5 status
red5 (pid 17731) is running...

Code:

[root@my.server.com:~]telnet localhost 5080
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.

Code:

[root@my.server.com:~]telnet localhost 1935
Trying 127.0.0.1...
Connected to localhost.
Escape character is '^]'.

Code:

[root@my.server.com:~]lsof -i:1935
COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME
java 17731 root 92u IPv6 3864989 TCP *:macromedia-fcs (LISTEN)
[root@my.server.com:~]

Code:

[root@my.server.com:~]lsof -i:5080
COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME
java 17731 root 100u IPv6 3864990 TCP *:5080 (LISTEN)
[root@my.server.com:~]

Hope you can give me any ideas about why it's not working on the port pass test. I attached the image of the failed test so you can take a look.

Thanks a lot!

MensaWater · 05-04-2011, 09:40 AM

Did you try the telnet tests from the remote host rather than the application? That is to say your issue may not be that the port isn't responding but rather something in the application.

telnet my.server.com 5080
telnet my.server.com 1935

Do those connect? Do they immediately drop?

I've seen issues where security in the application itself causes ports to drop even though they're open from a system perspective. Also is there any firewall device (switch/router/load balancer) between the systems? If so it may be that device that is blocking the traffic.

Finally it isn't exactly correct to say there is no firewall on the server - it is built into RHEL/CentOS. It may be correct to say you have it stopped. Run "iptables -L" to see what it is showing. If it is stopped there should only be 3 entries. Try running "service iptables start" then rerunning "iptables -L" to see if they are different. Try running "service iptables stop" then rerunning "iptables -L" to see if they are different.

sh4ka · 05-05-2011, 05:34 AM

Hi,

Thanks for your answer.

It is responding to telnet requests from outside. Also, there is no router/switch/balancer between this server.
iptables rules are empty, that's what I mean, even with the service started it shows this:

Code:

[root@myserver:~]iptables -L -n
Chain INPUT (policy ACCEPT)
target     prot opt source               destination         

Chain FORWARD (policy ACCEPT)
target     prot opt source               destination         

Chain OUTPUT (policy ACCEPT)
target     prot opt source               destination         
[root@myserver:~]

I'm starting to think it's an app problem...

Thanks.

MensaWater · 05-05-2011, 09:00 AM

I agree - it's probably the app. As noted in my prior post some apps have their own security settings that have to be modified to allow connections. (Oracle and Nagios are two I can think of off the top of my head.)