It looks like $my_mh is set by referencing a global var (ob_save) that looks like it is being set in the save_ob_msg.tcl that is also packaged in the recover 33 script. That whole process was in place prior to my tinkering.
As I’m writing this, I wonder if the issue is that I am sending these OB post-proc from SMAT. That seems like it would negate the save_ob_msg.tcl? I should probably try to resend them inbound and select only the OB thread that I want to try…