Samples with out of control points should be out of specifications.

verify, first your data and the conclusion.

Then work on whether the samples with 'out of control' (o.o.c.) points are in specs? or deserve to be rejected for bad specifications if any.

If out say Hoorrah.. and reject the lot

If out of control points /samples are still in specs; ignore those o.o.c. points and recalculate the control charts. If it is in control the points could be considered as sporadic. But be cautious for process parameters and monitoring.

if not.. ignore that and re-ascertain o.o.c. points.

As you have mentioned that you found some out of control data, it clearly indicates that your problem is to set the control limits for the process. This is called phase-I or the base period. In the base period , when the process shows points beyond control limits, it is required to find whether they are due to assignable causes or not. It requires to remove those points from the analysis. Then one goes into phase-II where the chart will be used to monitor the process.

Also, do not get confused with specification limits and control limits. Because I found some practioners are confused between USL and LSL with UCL and LCL. Please refer to Statistical quality conrol book by DCMontgomery for better understanding.

control limits are Statistical functions and as your are already plotting control chart - use of X-bar-R, X-bar-S, and X-MR , you seem to be conversant with USL and LSL

A process could be in control and yet not acceptable as a quality (fit) process. Alternatively, be out of control, yet be acceptable.

you need to be into the latter first, and then improve upon. [being conforming / fit is more important if one is consistent ].

So are you producing sales or rejections? (a specification problem?)

If you are in the preliminary stage ( that is phase 1) that means you are still training your data, so you can remove data points that led to the out of control until you observed in control process. So such data can now be used to monitor subsequent observations. While monitoring the subsequent observations, you are already at the stage two of your study, and what ever data points that goes out of the control limits is consider unfit for the process, it can be remove, rework, repair etc as the case may be.