Entire mainframe IPL, or just LPAR? My understanding is that usually you subdivide into few LPARs and then reboot the production ones on schedule to prevent dr

Entire mainframe IPL, or just LPAR? My understanding is that usually you subdivide into few LPARs and then reboot the production ones on schedule to prevent drift and ensure that yes, unplanned IPLs will work