We are facing a technical issue with networking, which I'm not able to comprehend.
We are using Linphone Client to connect to x.x.x.125(Freeswitch box) via x.x.x.101 i.e .101 being a SIP proxy Flexisip.
The SIP flow look as below.
[ Linphone box ] <-> [ `.101` box ] <-> [ `.125` box ]
Now when we register i.e SIP REGISTER request (without VPN connected) to .125 via .101 it works i.e Registration is a success, saying this because I was sniffing around the traffic with TCP dump on both .101 and Linphone Client box (both had 200 OK Response for Register request).
Now here is the deal, when we make a call i.e send SIP INVITE request without VPN connected. We see no traffic on .101 box but same can be found on Linphone box hinting that request is directed to .101.(but at the same time I can constantly see OPTIONS request appearing from .101 to Linphone Machine and also a 200K response getting sent from Linphone machine for OPTIONS request.)
Now, The sooner we connect to VPN, we see the request appearing on the .101 box from Linphone box
Now, had this behaviour would have stayed constant, I would have suspected the Firewall rule but the it would work during SIP REGISTER and not using Regular INVITE invite is something I'm able to understand here.
And when we are connected to VPN it work.
NOTE: If I assume this is a packet loss as a part of UDP even the retransmission too does not go through and this happen in multiple run..
It's only the INVITE packet would not be sent always does not happen with REGISTER request.
Here how the sip flow looks like
Given the REGISTER requests are getting through that rules out an IP routing and firewall issue (assuming it's not doing deep inspection on SIP packets).
That leaves the two most likely culprits as the client software (in this case Linphone) using the wrong network interface,
Try running a test with a tool like sipp where you can explicitly set the local address to use and the type of SIP request to send.
# To test the user agent client scenario (which sends INVITE requests) use:
sipp -bind_local 10.1.10.1 -sn uac -m 1 x.x.x.125
Update:
Some observations from inspecting the packet captures:
In the no VPN case:
there is a NAPT between the softphone and the Proxy translating 10.1.10.1 to 49.36.13.47, e.g. 10.1.10.1:39248 mapped to 49.36.13.47:44150.
REGISTER responses do appear to be coming from the FreeSWITCH server at 63.211.239.125 according to the User Agent string of Flexisip... on the response.
in agreement with the original post there are no responses at all to INVITE requests in the capture.
there are no fragmented UDP packets captured.
In the VPN case:
there is no NAPT between the softphone and the Proxy. The softphone traffic is originating from 172.17.8.37.
the softphone is using an IP address of 192.168.29.134 in it's SDP offer which means the device likely has multiple network interfaces.
there is a fragmented UDP packet recorded when the softphone sends an INVITE request to the Proxy but it doesn't seem to be an issue as the Proxy happily forwards the request to the FreeSWITCH server.
Missing Data:
The no VPN capture does not contain the traffic between the Proxy and the FreeSWITCH server. This is the most crucial leg for the analysis since it could show whether the Proxy is forwarding the INVITE request or not.
Running tcpdump directly on the Proxy would be able to provide this missing information.
Updated Guess:
Based on the still incomplete information my best guess would now be that that Proxy has misconfigured (or perhaps deliberate) SIP settings and is silently dropping INVITE requests received on public interfaces.
When the softphone connects on the VPN the INVITE requests are forwarded because they are considered to originate from an internal network.
For REGISTER requests the Proxy could have a rule that says always forward them no matter which interface the Proxy receives them on since they are not as risky as INVITES.
If the path is different, the INVITE will go through different networks and they may not behave the same.
I can see 2 possible issues with the network used when VPN is not active:
A NAT has an ALG which, being broken, would drop the INVITE and let the other ones go through. This is unlikely, because it's happening for several User-Agent your tried.
The network is configured to drop packet larger than a specific size. This is very likely because the INVITE, with all User-Agent, is always the largest SIP message being sent.
I would advise you:
Try TCP: This should confirm it's a UDP only issue.
Try to remove all codecs and keep only PCMA, with UDP: if it works, it's likely a UDP/MTU/SIZE issue.
EDIT:
To make it clear, you certainly have an MTU issue.
Thus, I advise you to test your UDP network and the MTU size limit with and without the VPN. You don't have to use your android, but you need to use the same network.
On sip server, starts:
$> nc -u -l -p 2399
On LAN side, any PC with same network with netcat tool...
$> cat invite1000.example | nc -u sip.antisip.com 2399
$> cat invite1200.example | nc -u sip.antisip.com 2399
$> cat invite1500.example | nc -u sip.antisip.com 2399
$> cat invite2000.example | nc -u sip.antisip.com 2399
$> cat invite8000.example | nc -u sip.antisip.com 2399
Make several invitexxx.example files with any data, but with specific number of char in it.
In theory, nc/netcat on the server will stop receiving the packets when going over the MTU (or the packet will be incomplete).
Then, this will confirm it's an MTU issue.
Related
Golang application with a client and server.
Server uses net.ListenUDP -- client also uses net.ListenUDP, connects to server and sends a packet with conn.WriteToUDP with the server address.
Server receives the packet with ReadFromUDP and grabs the return address. Using this return address, it then sends a large number of packets back to the client.
When running both client and server on local machine, this works perfectly. Using Wireshark I can inspect the UDP packets and see that they contain the source and destination ports - and in the application I can see that they arrive and my various checksum tests show the data is accurate.
I then moved the server off site to a remote machine. The application stops working. I can successfully send the first message from the client to server - this is received just fine. The server sends the response back 'toward' the client - but the client never receives them.
Using Wireshark, I can see that the packets do arrive back on the local machine with the correct IP address. It appears that my network router has performed NAT on the outgoing packets - and has correctly re-addressed response packets to the internal IP.. BUT there is no port.
So I have UDP packets arriving on the correct machine, but no port - so the client application does not receive them. Application times out on ReadFromUDP.
I don't know if it is relevant, but on local machine, Wireshark labels the packets as BT-uTP Utorrent packets. When they come in from remote server, this is what I see in Wireshark - note the lack of Port.
Any thoughts how I can solve this. I didn't think this was a UDP hole punching problem because although I am establishing a connection across a NAT it is with a server not a peer.
This packet is fragmented, You can see this under Internet Protocol Version 4 > Flags.
If you look at the frame as shown on the bottom of the picture you provided you should see the ports.
net.ListenUDP doesn't appear to support fragmentation at the socket level.
Do you have a PPPoe connection? You may need to reduce your packet size being sent by 8 bytes or change the MTU on the routers external interface of the remote side. You may also need to change the local routers MTU if it's on a PPPoe interface.
I have installed freeswitch on a machine which is kept in the LAN behind the firewall named PFSense.I was looking in the freeswitch logs and I found that there are some unknown sip request coming from the unknown IP from outside of the network.
I have not done any setting or port forwarding in the pfsense.But don't know why this is happening.Please tell me how can I block those requests.
Thanks
it's most probably just typical port scanners. They scan for port 5060 and then try to send some INVITE or REGISTER requests there. It's quite harmless if your PBX is secure enough.
Problem statement:
Suppose a parent server is hosted on a machine IP: 1.1.1.1 and that server some time communicates with three different servers say A (1.1.1.2), B (1.1.1.3), C (1.1.1.4). Those servers may be database servers or any other servers.
Now from your browser you can send a http request to 1.1.1.1/somePage.htm, as a result some TCP packet will go to the server 1.1.1.1, and 1.1.1.1 can send and receive some TCP packets from A,B,C as well.
Aim is to get the information of all TCP packets from the browser machine, without installing any agent software in any servers.
One solution is we can write a code at the 1.1.1.1 server machine that will filter all the TCP packets with respect to respective IPs. But I don’t want that solution.
Is there any way to solve this issue? Is it possible to introduce new protocol for this? But server codes can’t be modified.
Does "any agent software" includes something like Wireshark? Usually the way to look at all datagrams received is by using a sniffer like Wireshark or you can use tcpdump in Linux servers.
You can also use Netfilter to handle received packets in the server an take certain actions on them.
If all the above is included in what you don't want to do the only alternative I see is to add another server in the middle between the browser and the web server (or between the server and a load balancer if you have a load balancer) that acts only as a router or bridge. In that machine you can inspect and filter TCP segments with all the available tools.
I have chosen QuteCom SIP client for windows to chat.I have installed and configured the account with my public server. My SIP server is kamailio.The connection to the server is not established. The application is connecting to the server for a long time.
Any help is appreciated.
If looks like keep connecting, then I guess the SIP messages don't get to the server.
You can install Wireshark to monitor traffic on windows host on port 5060 (the SIP port) in order to see if SIP messages are sent to the server.
On server, you can install ngrep for the purpose of seeing if traffic from the phone comes there. The command would be like:
ngrep -d any -qt -W byline port 5060
If you don't see traffic coming to the SIP server, then might be a firewall or an ALG between the client and the server, or, a firewall even on client host or server itself.
If it is something in between (not on client host or server), then you should try to use TCP or better TLS.
Note that if you have the firewall on the server, you will see the SIP packets coming on the network, but they will be dropped by the kernel before getting to application layer. Typically on Linux you can see the firewall rules with:
iptables -L
If the SIP packets come to the server, then set debug=3 in kamailio.cfg, restart kamailio and watch the syslog file (e.g., /var/log/syslog or /var/log/messgaes) for kamailio-specific debug messages -- you should get hints of what happens during processing.
I currently have issues with SIP User Agents behind a symmetric NAT connecting to my SIP client, which is an IVR voice service.
I read that Asterisk has a solution for this in the sip.conf, where I can set attribute
nat=yes
and this will ignore the IP and Port in the SIP headers and use the one for the SIP request and also waits for an incoming RTP stream to reply to.
I'd like to make use of this feature as we already have an Asterix server installed for AIX requests.
What would be the minimum configuration required for Asterix to act as the man in the middle on a new port as 5060 will still be used to connect directly to the SIP client? I don't care about authentication etc. I just need the Asterix to act as a SIP relay.
Thanks
K
You may also need to have "canreinvite=no" in the [general] section of your sip.conf.
That setting keeps Asterisk in the call path, otherwise voice traffic may be sent directly from one endpoint to the other.