On Fri, 23 Feb 2007 12:46:21 -0600, Mark Zelden <[EMAIL PROTECTED]> wrote:
>So has anyone turned this trap on? Did you notice "a performance hit"? and >Why not just change the bleepin' default? We turned it ON coincident with the z/OS 1.8 roll out and have not experienced any noticeable performance degradation. It does seem like a good thing to enable since the type of corruption it prevents might not be detected until hours or days later. A version of the check actually shipped in the base 1.8 code. As I said in the SHARE pitch, it was poorly written and confusing. OA15539 cleans up the 1.8 version of the check and ships it for older releases. I tested a ++APAR version of this and the exception messages look much better. There is still a timing issue at IPL though, and if you start the HC SUB=MSTR, you will have to DELETE the check (or make it INACTIVE) in order to avoid getting an exception. As many IBMers know, I certainly agree with Mark that defaults should not be flagged as exceptions. Instead the default should be changed to the "best practice" setting. In talking with VSAM L2, I think there's a good chance that the index trap will default to ON in the future. On Sun, 25 Feb 2007 17:45:10 -0500, Bob Rutledge <[EMAIL PROTECTED]> wrote: >Before this discussion, did anyone even know this trap had existed for three years or was I the only ignorant soul? Unless apar OA03570 caught your eye in Jan. 2004 (I assume the PTFs had ++DOC holds) I don't know how you would have known about this. I sure didn't. So in that respect, the health check has value in that it makes people aware of the trap. -Dave ---------------------------------------------------------------------- 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
