What's the way out of this problem? Can the single-crystal technique do this task? If yes, What if we also can't make a single crystal of this powder material?
I suggest you first confirm the sample is close to ideal powder form (e.g. randomly oriented, strain-free, particle size between 1-5 um, etc.), the sample is really single-phase or a mixture of known phases, and the X-ray diffraction setup is appropriate.
After confirming the experiment is appropriate and the phase is unreported in the literature, you may consider if solving the crystal structure of the unknown phase is of your research interest and conceive the experiment accordingly.
After following Bill Chang's advice, if you do need to determine the space group, then make sure you have high quality data that starts at sufficiently low angle to get the first peaks, then you would index the pattern. Google indexing diffraction pattern or follow tutorials for using your instrument's software for that function.