> On May 16, 2017, at 10:40 AM, Walt Farrell <[email protected]> wrote: > > On Tue, 16 May 2017 09:57:16 -0400, Sam Golob <[email protected]> wrote: > >> That having been said, the system doctor sometimes has to deal with >> things that go wrong. It's nice when the system is working as >> designed. But sometimes, the NON-CANCELABLE job or STC goes awry, and >> it has to be restarted. In such a case, as in the middle of a day's >> production, you want to avoid an emergency IPL. And so you need a tool >> in the toolbox, to cancel the job or STC. Sometimes the only solution >> is to blow it away. > > However, you should be prepared, when you use the tool, to have to IPL > anyway. And that should be clearly stated in any documentation for the tool. > > -- > Walt —————————SNIP———————————
In the early days of MVS we had constant issues of jobs going non cancelable. Our IBM SE wrote a super cancel (callrtm) command. We *WERE* using it sparingly. It was a last grasp to IPL. Although most of our issues had to due with allocation (Q4) getting hung. We were IPLing once or twice in 3 days (sometimes as many as 8). IBM rewrote allocation and just about eliminated the need for it. We did still use it from time to time and it did save IPL’s. Our other IBMer who worked behind mountains of standalone dumps was used to it and didn’t raise it as an issue. But it was understood by the group that using it was a last gasp attempt and then only making sure there were no allocation hangups. Ed ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
