Re: [SR-Users] Rtpengine vs. TURN?

2014-07-25 Thread Daniel-Constantin Mierla
Hello, On 14/07/14 15:49, Peter Villeneuve wrote: Hi Daniel, Thanks for your input. Since I couldn't decide which one to use, I've been experimenting with using both. The problem with my mixed approach is that there are too many ICE candidates created (I counted 10 in the last logs I looked

Re: [SR-Users] Rtpengine vs. TURN?

2014-07-14 Thread Daniel-Constantin Mierla
Hello, On 12/07/14 19:55, Peter Villeneuve wrote: Hi, On my server, I have the option of using either Rtpengine for NAT traversal or pure TURN without rtpengine. Rtpengine has the obvious plus that it only needs 1 public IP, while TURN (with STUN) will need 2 public IPs, although that's not

Re: [SR-Users] Rtpengine vs. TURN?

2014-07-14 Thread Peter Villeneuve
Hi Daniel, Thanks for your input. Since I couldn't decide which one to use, I've been experimenting with using both. The problem with my mixed approach is that there are too many ICE candidates created (I counted 10 in the last logs I looked at for one call), real relay candidates (turn), and

[SR-Users] Rtpengine vs. TURN?

2014-07-12 Thread Peter Villeneuve
Hi, On my server, I have the option of using either Rtpengine for NAT traversal or pure TURN without rtpengine. Rtpengine has the obvious plus that it only needs 1 public IP, while TURN (with STUN) will need 2 public IPs, although that's not a problem in my case. Having said that, I'd like to