Ron Hawkins wrote: > I have not heard of a CA being "cleared out." My old notes say that all > CI with a key higher than the mid-point sequence set record are written > to the new CA. The space in the old CA is now marked free - it is not > formatted.
Au contraire, Ron. On a CA split, after moving half of the CIs from the source CA, it's necessary for VSAM to format the remaining CIs in the "other half" of the target CA. Additionally, VSAM must then reformat all of the "moved" CIs in the source CA. Otherwise, the emergency capability to read the KSDS as an ESDS (i.e., without using the Sequence Set to read it in key sequence) in the event the index is blown would not be possible. If VSAM didn't format, just imagine how good MVS (yup, when I wrote the VSAM book, it was still called MVS) data security would be if you read that KSDS as if it was an ESDS and those disk tracks in the target CA just happened to have old "deleted" data laying around on them that somehow "looked" like your VSAM data set (same 4K CISZ, which many people use, for example); also, if VSAM didn't reformat the original CIs, you'd physically have duplicate data on those tracks with the data that's now out in your target CA. BTW, Bob (and congratulations on your new job), winter is not a real good time to be off the west coast of Canada in small craft, even if you're on the Inside Passage. My wife and I are heading to SE Alaska in early May, though, and will probably be out for four months. We're cruising these days on a 42' Nordic Tug (www.nordictugs.com), built locally here in good 'ole Anacortes, WA. And no, the VSAM reference book doesn't go with me. Take care everyone. Ron Ferguson ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
