Allen, thanks for taking the time to explain. I'll look into MINDORM/MAXHOLD, the BROADCAST dsn is something I have wanted to change for a while. Maybe now is a good time to finally do something.
Thanks Matt -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Staller, Allan Sent: Tuesday, February 18, 2014 1:22 PM To: [email protected] Subject: Re: IEF196I Message Not arguing that point. All of this conversation stems from the Other Image attempting to contact device 4299 and did not get a response in a timely manner. A "timely manner" is defined by the MIH values in IECIOSxx on the "other system". This is mostly a good thing, but carried to extremes causes this issue. 1) for whatever reason, MVS1 indicates to other images that device 4299 is busy. This can come from a variety of sources. 2) The other images MIH is waiting for a response. If this response is not received, the MIH causes the Start Pending message to be issued. From the viewpoint of MVS1: 1) JESCKPT (if not tuned), hangs a never-ending reserve on the CHKPT volume. This will cause the other systems to see a continuous device busy and eventually, MIH will trip and the Start Pending message will be issued on the other image. 2) As previously noted, SMF19 processing will attempt to contact the device, regardless of the online/offline status. If the device is marked busy either by reserve processing or just general activity, this will also cause the Start Pending. Normally this will occur at startup/shutdown time. 3) BROADCAST uses a very old technology and is a known contributor to excessive Device Busy. In short, when the device is *very* busy to MVS1, the OTHER SYSTEM(s) will tend to generate start pendings. The solution is to reduce the DEVICE Busy on MVS1 so that it can respond to the other image in a "timely" manner. The other choice is to "live with it". As indicated above, my plan of attack would be to adjust the MINDORM/MAXHOLD values in the JES CKPTDEF, followed by relocating datasets and then the SMF processing. HTH, </snip> Allen, this volume is supposed to be online and used to the lpar you asked about. It's the sandbox where the IOS071I 4299,**,*MASTER*, START PENDING 680 occurred. That is where the volumes is offline and not used but where the message appeared. Thanks Matt <snip> ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
