On Tue, 25 Jun 2013 15:47:09 -0500, Shane Ginnane <[email protected]> wrote:
>I'm was surprised to see GRS utilising XCF still on ESCON these days, but ring >is still out there. In this town of less than a handful of sites, one recently >had to do the ESCON to FICON in a hurry to get a couple of EC12s in the door, >and another isn't in a position to take an outage to do the change at the >moment. If all goes according to Hoyle, no outage is necessary. If the hardware or z/OS release does not support FICON, or they are not enabled for dynamic I/O reconfiguration (I haven't seen these conditions in a decade or more, but I can believe it's still out there), then clearly an outage (or three) will be needed. If they are in a mixed GRS complex, the documentation provided by others in this thread will spell out the requirements. Plan, then define the new CTC links to HCD and ACTIVATE the updated config. SETXCF START PATHIN/PATHOUT commands to allocate and start transmitting over the new devices. SETXCF STOP PATHIN/PATHOUT commands to stop traffic on the old devices. Unless I'm missing something obvious (been a few years since the migration at my last shop), and Bob's your uncle and Fanny's your aunt. Oh, yeah...remember to update PARMLIB for the next IPL. Migration to STAR is a little harder if you don't already have a CF for the Sysplex, or if you don't have enough storage to accommodate the ISGLOCK structure, otherwise, this too can be done dynamically (fallback sucks though). Cheers, Art ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
