That seems to be an INDI OnStep driver bug.
 
I don't use INDI nor am I involved in its OnStep Telescope driver development, but looking at the source code:
https://github.com/indilib/indi/blob/master/drivers/telescope/lx200_OnStep.cpp
 
Line 4674 (at the moment) needs to look like this:
    if(error_or_fail != 4 || read_buffer[0] < '0' || read_buffer[0] > '9' || read_buffer[1] < '0' || read_buffer[1] > ':'

// :A?#       Align status
    //            Returns: mno#
    //            where m is the maximum number of alignment stars
    //                  n is the current alignment star (0 otherwise)
    //                  o is the last required alignment star when an alignment is in progress (0 otherwise)
    if (command[1] == '?' && parameter[0] == 0) {
      reply[0] = '0' + ALIGN_MAX_NUM_STARS;
      reply[1] = '0' + alignState.currentStar;
      reply[2] = '0' + alignState.lastStar;
      reply[3] = 0;
      *numericReply = false;

#define ALIGN_MAX_STARS              AUTO //   AUTO, Uses HAL specified default (either 6 or 9 stars.)                        Infreq
                                          //         Or use n. Where n=1 (for Sync only) or 3 to 9 (for Goto Assist.)

mistake is believing that alignState.currentStar can only assume values up to 9.  It can be 10 in 9-star align mode and being one greater than the number of alignment stars is the condition that signals the alignment is done.
 
So in the code '0' + 10 == ':' which is a bit of an ugly hack but that's how it's done.