Re: System Free Storage
A check of Q RECORDING Looks OK, we're only doing ACCOUNT, no COUNT. I'm pretty much going to pass this whole thread over to the owner of the query machines; It is to my knowledge a recent development and from my vantage, there have been recent changes in the performance/resource usage of them. Rob, would something show in the ESA R MENU about what is a constrained CP storage? I confess to not fully understanding ESASXS but ESAFREE shows free storage pages, long term around 1700, rest of the cols are zeros... but that is lower than other VM LP's with less storage... we've 22720M central/4032M expanded. The machines do about as many IOs and IUCVs Userid: CPU% IO/s DASDio IUCV/s Diag98 BlkI/O PG/s WSS(m) Share/CP 1)SWEEP4 11.1 40.9 40.9 30.4 0.0 0.0 0 8.7 100% 2 line(s) not displayed 4)SWEEP3 5.3 28.0 28.0 20.3 0.0 0.0 0 2.4 100% 5)SWEEP1 4.5 25.5 25.5 18.7 0.0 0.0 0 2.4 100% 6)SWEEP2 3.5 20.4 20.4 14.6 0.0 0.0 0 2.4 100% and maybe I'm not looking at the right place in the VMDBK/my calc's... SWEEP1-4, I'm pm0gar and I picked up the 8E'x by invoking TASTE app and PF3ing out; about B'x/C'x per App/PF3 09:38:21 PM0GAR 008E'x142'd DWs 0 Pages of free storage. 09:38:21 SWEEP1 00091377'x 594807'd DWs 1161 Pages of free storage. 09:38:21 SWEEP2 009D4FB2'x 10309554'd DWs 20135 Pages of free storage. 09:38:21 SWEEP3 00D820F2'x 14164210'd DWs 27664 Pages of free storage. 09:38:21 SWEEP4 003D4B00'x4016896'd DWs 7845 Pages of free storage. 09:39:21 PM0GAR 008E'x142'd DWs 0 Pages of free storage. 09:39:21 SWEEP1 000915F3'x 595443'd DWs 1162 Pages of free storage. 09:39:21 SWEEP2 009D5002'x 10309634'd DWs 20136 Pages of free storage. 09:39:21 SWEEP3 00D821EB'x 14164459'd DWs 27664 Pages of free storage. 09:39:21 SWEEP4 003D4F12'x4017938'd DWs 7847 Pages of free storage. 09:40:21 PM0GAR 008E'x142'd DWs 0 Pages of free storage. 09:40:21 SWEEP1 00091843'x 596035'd DWs 1164 Pages of free storage. 09:40:21 SWEEP2 009D5156'x 10309974'd DWs 20136 Pages of free storage. 09:40:21 SWEEP3 00D82365'x 14164837'd DWs 27665 Pages of free storage. 09:40:21 SWEEP4 003D5333'x4018995'd DWs 7849 Pages of free storage. 09:41:21 PM0GAR 008E'x142'd DWs 0 Pages of free storage. 09:41:21 SWEEP1 00091A83'x 596611'd DWs 1165 Pages of free storage. 09:41:21 SWEEP2 009D520F'x 10310159'd DWs 20137 Pages of free storage. 09:41:21 SWEEP3 00D824AD'x 14165165'd DWs 27666 Pages of free storage. 09:41:21 SWEEP4 003D5706'x4019974'd DWs 7851 Pages of free storage. On Wed, Dec 16, 2009 at 3:54 PM, Rob van der Heij rvdh...@gmail.com wrote: Just wagging, but the fair share before CP gets annoyed was calculated as a part of available storage. If the available storage gets very low, even fairly normal usage is considered outrageous by CP. The case that I had was where pending account or EREP records were filling up available memory and CP even felt that OPERATOR doing an IPL CMS was enough to kick him off... This was in the old days, so I don't know what all resides in CP free storage address space. Rob -- Gregg Reed No Plan, survives execution
Re: System Free Storage
Alan, Thanks. Probably after the 1st but I'll discuss opening a PMR sooner with our VM guys and the machine(s) owner. I don';t think we'll see a FRF002 as the machines have been stopped/forced. I'm not planning on them setting STREXEMPT for the machines or STRLIMIT off either! Happy Holidays On Wed, Dec 16, 2009 at 1:45 PM, Alan Altmark alan_altm...@us.ibm.com wrote: You need to open a z/VM PMR so that the Support Center can work with you to determine what is triggering those messages. They occur because your fair share of certain control blocks have been consumed. But because CP doesn't operate with formal threads, the free storage monitor can't readily unwind the memory request to find out *why* the control block was needed. They can be consumed by instruction simulation, diagnose instructions, IUCV, commands invoked via diag 8, excessive console I/O, etc. Alan Altmark z/VM Development IBM Endicott -- Gregg Reed No Plan, survives execution
Re: System Free Storage
Gregg, Can you give us a general idea of what the machines are supposed to be doing? Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gregg Sent: Wednesday, December 16, 2009 9:25 AM To: IBMVM@LISTSERV.UARK.EDU Subject: System Free Storage We've an application that gets HCPLIM6007I, and eventually 6008E or 6009E Short of logging off/on, is there a way for the logon to advise CP they nolonger need it? It seems to survive a CP IPL or CP SYSTEM CLEAR... The disconnected machines, don't appear to be doing any of the items mentioned under User Response in HCP6007I, but I do not have access to the source and they maybe issuing CP commands that consume free storage... aside from getting the source and trying any, is there any way to know ahead of time what CP commands those might be? -- Gregg Reed No Plan, survives execution
Re: System Free Storage
Richard, they are sending performance queries to TPF images; ACTIVE(ACTION? can't recall for sure), VISTA like but called TASTE on this system over ESCON CTCs; We've an IP flavor that we're moving to that doesn't have this issue or hasn't shown any in other systems. Many years ago, our VISTACTL machine had a similiar problem but I don't recall any of the query machines having it. On Wed, Dec 16, 2009 at 12:28 PM, Schuh, Richard rsc...@visa.com wrote: Can you give us a general idea of what the machines are supposed to be doing? -- Gregg Reed No Plan, survives execution
Re: System Free Storage
-Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gregg Sent: Wednesday, December 16, 2009 11:25 AM To: IBMVM@LISTSERV.UARK.EDU Subject: System Free Storage We've an application that gets HCPLIM6007I, and eventually 6008E or 6009E Short of logging off/on, is there a way for the logon to advise CP they nolonger need it? It seems to survive a CP IPL or CP SYSTEM CLEAR... The disconnected machines, don't appear to be doing any of the items mentioned under User Response in HCP6007I, but I do not have access to the source and they maybe issuing CP commands that consume free storage... aside from getting the source and trying any, is there any way to know ahead of time what CP commands those might be? -- Gregg Reed Just out of curiousity, have you tried doing a CP DEFINE STORAGE? -- John McKown Systems Engineer IV IT Administrative Services Group HealthMarkets(r) 9151 Boulevard 26 * N. Richland Hills * TX 76010 (817) 255-3225 phone * (817)-961-6183 cell john.mck...@healthmarkets.com * www.HealthMarkets.com Confidentiality Notice: This e-mail message may contain confidential or proprietary information. If you are not the intended recipient, please contact the sender by reply e-mail and destroy all copies of the original message. HealthMarkets(r) is the brand name for products underwritten and issued by the insurance subsidiaries of HealthMarkets, Inc. -The Chesapeake Life Insurance Company(r), Mid-West National Life Insurance Company of TennesseeSM and The MEGA Life and Health Insurance Company.SM
Re: System Free Storage
Hadn't but just tried; 16M machine, CP DEF STOR 16M and STOR 32M; still have, 142 double words if I'm looking at the right place in the VMDBK... and I think I am... Just out of curiousity, have you tried doing a CP DEFINE STORAGE? -- John McKown Gregg Reed No Plan, survives execution
Re: System Free Storage
Vista. That brings back memories. Howard Grams created the original ACTIVE while at TWA. Vista was our second generation, much more versatile, replacement, written at Piedmont Airlines in 1985. Howard wrote the code for the VISTACTL machine and the ALC terminal emulation; I wrote the CTC code. The same text deck was used in both the TPF and CMS machines. The only CP Free Storage used for the CTC connection was whatever was needed to support the VCTCAs. The rest of its storage was buffers in the various machines. IIRC, the CMS machines communicated with VISTACTL using SMSGs, with large messages being broken down into appropriate sizes and reassembled by VISTACTL before being sent to the TPF machines over the CTC connection. The return messages, depending on their sizes, were subject to the same process. That is a more likely area of concern, depending on the number and size of the queries and responses. It may be that the smsgs were piling up faster that they were being retrieved. I/P was our next goal; however, outsourcing ended that. Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gregg Sent: Wednesday, December 16, 2009 9:40 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: System Free Storage Richard, they are sending performance queries to TPF images; ACTIVE(ACTION? can't recall for sure), VISTA like but called TASTE on this system over ESCON CTCs; We've an IP flavor that we're moving to that doesn't have this issue or hasn't shown any in other systems. Many years ago, our VISTACTL machine had a similiar problem but I don't recall any of the query machines having it. On Wed, Dec 16, 2009 at 12:28 PM, Schuh, Richard rsc...@visa.com wrote: Can you give us a general idea of what the machines are supposed to be doing? -- Gregg Reed No Plan, survives execution
Re: System Free Storage
On Wednesday, 12/16/2009 at 12:25 EST, Gregg reed.gr...@gmail.com wrote: We've an application that gets HCPLIM6007I, and eventually 6008E or 6009E Short of logging off/on, is there a way for the logon to advise CP they nolonger need it? It seems to survive a CP IPL or CP SYSTEM CLEAR... The disconnected machines, don't appear to be doing any of the items mentioned under User Response in HCP6007I, but I do not have access to the source and they maybe issuing CP commands that consume free storage... aside from getting the source and trying any, is there any way to know ahead of time what CP commands those might be? You need to open a z/VM PMR so that the Support Center can work with you to determine what is triggering those messages. They occur because your fair share of certain control blocks have been consumed. But because CP doesn't operate with formal threads, the free storage monitor can't readily unwind the memory request to find out *why* the control block was needed. They can be consumed by instruction simulation, diagnose instructions, IUCV, commands invoked via diag 8, excessive console I/O, etc. Alan Altmark z/VM Development IBM Endicott
Re: System Free Storage
The message says that it is CP Free Storage that is being depleted, not virtual storage. Defining storage is not likely to be a solution. Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Gregg Sent: Wednesday, December 16, 2009 9:51 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: System Free Storage Hadn't but just tried; 16M machine, CP DEF STOR 16M and STOR 32M; still have, 142 double words if I'm looking at the right place in the VMDBK... and I think I am... Just out of curiousity, have you tried doing a CP DEFINE STORAGE? -- John McKown Gregg Reed No Plan, survives execution
Re: System Free Storage
Correct, though redefining virtual machine storage to a different size performs a reset clear function on the virtual machine, which could well help clear up the problem. But as Alan says, this isn't typically something a customer can unravel looking at the system from the outside. - Bill Holder, z/VM Development, IBM On Wed, 16 Dec 2009 10:47:49 -0800, Schuh, Richard rsc...@visa.com wrot e: The message says that it is CP Free Storage that is being depleted, not virtual storage. Defining storage is not likely to be a solution. Regards, Richard Schuh
Re: System Free Storage
Ok, I should have read the whole thread more carefully - both re-IPL and SYSTEM CLEAR had already been tried with no effect, redefining virtual storage would be a similar no-op. I think Richard's probably on the righ t track with the theory about SMSG backlogs building up, resetting the virt ual machine wouldn't clear that, I don't think. - Bill Holder, z/VM Development, IBM On Wed, 16 Dec 2009 12:57:40 -0600, Bill Holder hold...@us.ibm.com wrot e: Correct, though redefining virtual machine storage to a different size performs a reset clear function on the virtual machine, which could well help clear up the problem. But as Alan says, this isn't typically something a customer can unravel looking at the system from the outside. - Bill Holder, z/VM Development, IBM
Re: System Free Storage
Simply a SYSTEM CLEAR or an IPL CLEAR would work as well. So would a LOG OFF/ON sequence. None is an actual solution of the problem, they simply buy time by sweeping it under the rug until the next time. If I had a recurring problem of CP Free Storage being consumed to its limits, I would want to find a real solution before the system failed because of it. Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:ib...@listserv.uark.edu] On Behalf Of Bill Holder Sent: Wednesday, December 16, 2009 10:58 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: System Free Storage Correct, though redefining virtual machine storage to a different size performs a reset clear function on the virtual machine, which could well help clear up the problem. But as Alan says, this isn't typically something a customer can unravel looking at the system from the outside. - Bill Holder, z/VM Development, IBM On Wed, 16 Dec 2009 10:47:49 -0800, Schuh, Richard rsc...@visa.com wrot= e: The message says that it is CP Free Storage that is being depleted, not virtual storage. Defining storage is not likely to be a solution. Regards, Richard Schuh
Re: System Free Storage
Bill Holder wrote: Ok, I should have read the whole thread more carefully - both re-IPL and SYSTEM CLEAR had already been tried with no effect, redefining virtual storage would be a similar no-op. I think Richard's probably on the right track with the theory about SMSG backlogs building up, resetting the virtual machine wouldn't clear that, I don't think. Hmm.. Wouldn't a system reset clear any VMCF Authorize (for SMSGs received with SMSG ON) or IUCV path to *MSG (for SMSGs received with SMSG IUCV) - thus throwing down the drain any pending SMSG ? --Ivan
Re: System Free Storage
On Wed, Dec 16, 2009 at 7:57 PM, Bill Holder hold...@us.ibm.com wrote: Correct, though redefining virtual machine storage to a different size performs a reset clear function on the virtual machine, which could well help clear up the problem. Just wagging, but the fair share before CP gets annoyed was calculated as a part of available storage. If the available storage gets very low, even fairly normal usage is considered outrageous by CP. The case that I had was where pending account or EREP records were filling up available memory and CP even felt that OPERATOR doing an IPL CMS was enough to kick him off... This was in the old days, so I don't know what all resides in CP free storage address space. Rob