Alan, I went to talk to the z/os guy to get a better understanding of what the TCP IP folks are wanting done.
I now have a better understanding of what is going on. 1. We have so much data flowing through the hipersocket chipid that we are experiencing queueing in the hipersocket. (This is on a z196.) 2. It appears that something, either hipersocket or TCPIP, is not reacting well to that queueing. It appears that is what is causing the storage creep. 3. One of the things that the TCP IP folks want us to do on the hardware side is to increase the maximum frame size to 64k. for this hipersocket. 4. This would normally increase the MTU size to 56k for the hipersocket. 5. If I remember properly the 56k value would exceed the SAP recommended MTU size of 8k. 6. The IBM z/os TCP/IP support says that the MTU size could be forced to 8k. Anyone have any experience in forcing an MTU to 8k like this? Any unwelcome side effects? Thanks, Ron Ron Foster Baldor Electric Company 5711 R S Boreham Jr Street Fort Smith, AR 72901 Phone:479-648-5865 Fax:479-646-5440 Email: ron.fos...@baldor.abb.com IM Address:rfos...@baldor.com www.baldor.com ________________________________________ From: Linux on 390 Port [LINUX-390@VM.MARIST.EDU] on behalf of Alan Altmark [alan_altm...@us.ibm.com] Sent: Thursday, June 13, 2013 3:29 PM To: LINUX-390@VM.MARIST.EDU Subject: Re: Appropriate hipersocket MFS size for use with SAP Apps Servers On Thursday, 06/13/2013 at 03:29 EDT, Ron Foster <ron.fos...@baldor.abb.com> wrote: > We have a bunch of SAP application servers running under z/vm. They are all > pointed at the same z/os 1.13 system. There are several DB2s running on this > z/OS system. All the Apps servers are using the same hipersocket chpid to > communicate to the z/os system the same hipersocket chipid. Since going to z/OS > 1.13, we are experiencing storage creep in our TCPIP address space.> > > This hipersocket is using a 8k MTU size. I believe this to be the SAP > recommended size. > > Our z/os support folks are being told by their support folks that they would > like us to increase the MTU size from 8k to something larger. That sounds strange, to say the least. What logic do they give to support the conclusion that a larger MTU will prevent storage creep? Large MTU or small, the path through the stack should be the same. If you didn't have storage creep on the earlier release with an 8K MTU, you shouldn't have it now. Alan Altmark Senior Managing z/VM and Linux Consultant IBM System Lab Services and Training ibm.com/systems/services/labservices office: 607.429.3323 mobile; 607.321.7556 alan_altm...@us.ibm.com IBM Endicott ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For more information on Linux on System z, visit http://wiki.linuxvm.org/ ---------------------------------------------------------------------- For LINUX-390 subscribe / signoff / archive access instructions, send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit http://www.marist.edu/htbin/wlvindex?LINUX-390 ---------------------------------------------------------------------- For more information on Linux on System z, visit http://wiki.linuxvm.org/
