Recovering from data CI damage

REPRO. the data component of the cluster. Specify the data component name (not the cluster name) in the REPRO INFILE parameter.
If the REPRO fails, you will have copied all the available data that precedes the damaged portion of the data set. The high-used RBA value in the catalog entry for the data set that you specify in the REPRO OUTFILE parameter indicates how many bytes were successfully recovered before the damage was encountered. To recover valid data following the damaged CI, add the CISIZE (from the LISTCAT of the damaged data set) to the high-used RBA of the OUTFILE. Copy from the damaged data set data component using the REPRO FROM parameter. Copy FROM the high-used RBA + CISIZE, effectively skipping the damaged CI.
Continue this process, skipping damaged CIs until you have retrieved all valid data.
Note: The extent of the damage affects how much data you can successfully recover. If a backup copy of the data set exists, it may contain more valid data than the data set that results from this recovery method.