satoruhiga / ofxnatnet Goto Github PK

I'm getting an error when streaming two rigid bodies in the Example project (streaming one rigid body works perfectly, btw). Here's the output:
example(10915,0x7000002a0000) malloc: *** mach_vm_map(size=18446744073388515328) failed (error code=3) *** error: can't allocate region *** set a breakpoint in malloc_error_break to debug
Here's a screenshot during debugging:

Running Motive Tracker 1.9.0 and NatNet 2.9 on OSX 10.11.4

When running Natnet2OSCbridge using ofxNatNet on windows 10 I get an error:
[ error ] ofxNatNet: Net Exception

And therefor cannot connect to NatNet/Motive.
Using Motive version 1.8.0 and NatNet version 2.8.0.0

Using latest build with OF 0.9.8 and ofxNatnet master of github (https://github.com/satoruhiga/ofxNatNet)

Further research leads to locating the error in ofxNatNet.cpp starting on line 120 in the following try/catch:

try{ interface = Poco::Net::NetworkInterface::forAddress(Poco::Net::IPAddress(interface_name)); } catch (const Poco::Net::InvalidAddressException& e) { interface = Poco::Net::NetworkInterface::forName( interface_name, Poco::Net::NetworkInterface::IPv4_ONLY); }

The problem is that the Poco library returns a malformed IPV6 address instead of a IPV4 address.

WORKAROUND
A workaround is to turn of IPV6 in you network adapter settings and then it will work fine.

NOTE
This bug could be because of the use of static IPV4 IP addresses on the NATNET data network without any DHCP or DNS services.

We ran into an issue that we think is caused by the "copy to mainthread" code of a MessageID == 7 packet (NatNet 2.8, Motive 1.8)

In the loop rigidbodies[i] is copied to this->rigidbodies[RB.id], which causes a jumble to happen when you do the following:

• Add 3 rigidbodies
• Remove rigidbody 1
• Add another rigidbody (which appears to receive id = 1 from Motive)

In this case we received the data in the order: 2, 3, 1, but the names in the order 1, 2, 3 (or the other way around, it's confusing)

Fix appears to be easy, change this->rigidbodies[RB.id] to "this->rigidbodies[i]" (works for us so far).