I can't test this, so please speak up if it solves things for you. Case in point: ekg/vcflib#11
Sorry for the delay, and thank you very much for the help!
A failed setRegion() leaves a ti_iter which shouldn't be destroyed. Attempting to call ti_iter_destroy on it in the next call to setRegion() will cause a segfault. Reordering the calls avoids this issue.