Re: CRR Machine
Richard, I restarted our CRR machine last weekend to upgrade it to CMS23 from CMS22. It did not hurt the other SFS SVMs while down. I cannot say whether it will interface again with the other SFS SVMs when it is brought back up because when I brought it back up, I brought the others down to upgrade them too. Jim -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Schuh, Richard Sent: Wednesday, November 28, 2007 12:10 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: CRR Machine We have no protected conversations, so it is more the reference to serious performance degradation when running in limp mode that may affect us. The questions of whether CRR could be stopped without causing the file pool servers to crash, and whether restarting the CRR machine while the file pool servers were running would restore normal operation were both addressed by Kris in an earlier post. In the absence of dissenting opinion, it appears that I will be able to move the CRR machine's disks with little or no noticeable effect, so long as I choose a lightly loaded period in which to do it. Regards, Richard Schuh -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Alan Altmark Sent: Wednesday, November 28, 2007 7:57 AM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: CRR Machine On Tuesday, 11/27/2007 at 11:39 EST, Alan Ackerman [EMAIL PROTECTED] wrote: The question is, when are two-phase commits used? From what others have told you, one answer is when you have R/W access to two different SFS filepools. The reference to a protected conversation is an APPC LU 6.2 SYNCLEVEL=SYNCPOINT conversation. It has a two-phase commit semantic built into it. Within a single CMS workunit you can, for example, open an SFS file (one file in one server, or multiple files, or multiple servers) and talk to one or more CICS LU 6.2 transactions (for example) over the network. When your application or one of the CICS transactions COMMITs the workunit, all files and transactions are committed or all are backed out. It doesn't matter whether you're doing this explicitly or not. The CMS file system will try to connect to the CRR server in *anticipation* of your opening a second file or establishing a syncpoint connection. Alan Altmark z/VM Development IBM Endicott This message w/attachments (message) may be privileged, confidential or proprietary, and if you are not an intended recipient, please notify the sender, do not use or share it and delete it. Unless specifically indicated, this message is not an offer to sell or a solicitation of any investment products or other financial product or service, an official confirmation of any transaction, or an official statement of Merrill Lynch. Subject to applicable law, Merrill Lynch may monitor, review and retain e-communications (EC) traveling through its networks/systems. The laws of the country of each sender/recipient may impact the handling of EC, and EC may be archived, supervised and produced in countries other than the country in which you are located. This message cannot be guaranteed to be secure or error-free. This message is subject to terms available at the following link: http://www.ml.com/e-communications_terms/. By messaging with Merrill Lynch you consent to the foregoing.
Re: CRR Machine
No problem for the SFS servers. There would be a problem for CMS end-users: they will not be able to open files in RW located in more than 1 SFS server at the same time. But, as soon as CRR is back, all works fine again. 2007/11/27, Schuh, Richard [EMAIL PROTECTED]: Can the CRR machine be taken down for s brief interval without taking down other SFS servers, or is it necessary to stop all SFS activity to do it? The situation is that the disks of the CRR machine are on physical disks that are to be decommissioned and I need to move them to a more permanent location. Naturally, if this were possible, it would take place during a period of extremely low activity. Regards, Richard Schuh -- Kris Buelens, IBM Belgium, VM customer support
CRR Machine
Can the CRR machine be taken down for s brief interval without taking down other SFS servers, or is it necessary to stop all SFS activity to do it? The situation is that the disks of the CRR machine are on physical disks that are to be decommissioned and I need to move them to a more permanent location. Naturally, if this were possible, it would take place during a period of extremely low activity. Regards, Richard Schuh
Re: CRR Machine
Thanks, Kris. Our users cannot open files R/W in more than 1 file pool at a time. We only have 2 pools, and R/W access to one of them is limited to a few service machines that have no authority in the other. Similarly, the general users enrolled in the other pool cannot write in the one being serviced by the few service machines. All I have to be concerned about is the FTP server. I can halt it for the few minutes required to move the disks. Regards, Richard Schuh From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Kris Buelens Sent: Tuesday, November 27, 2007 1:11 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: CRR Machine No problem for the SFS servers. There would be a problem for CMS end-users: they will not be able to open files in RW located in more than 1 SFS server at the same time. But, as soon as CRR is back, all works fine again. 2007/11/27, Schuh, Richard [EMAIL PROTECTED]: Can the CRR machine be taken down for s brief interval without taking down other SFS servers, or is it necessary to stop all SFS activity to do it? The situation is that the disks of the CRR machine are on physical disks that are to be decommissioned and I need to move them to a more permanent location. Naturally, if this were possible, it would take place during a period of extremely low activity. Regards, Richard Schuh -- Kris Buelens, IBM Belgium, VM customer support
Re: CRR Machine
Are the warnings about Limp Mode if there is no CRR machine still extant? If so, there is still something being done by the CRR machine. When lightly loaded, I would not expect it to be a problem. Regards, Richard Schuh From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Michel Raicher Sent: Tuesday, November 27, 2007 1:20 PM To: IBMVM@LISTSERV.UARK.EDU Subject: Re: CRR Machine The CRR machine only get used when a user access and write to 2 different sfs's at same time on the same work unit. This is not likely to be your case, so shoud be safe to stopped it. When the CRR machine is not there, when a user try to coordinated writing to 2 sfs's at same work unit, it will fail, like I said it is not likely you are doing this, regards Inactive hide details for Schuh, Richard [EMAIL PROTECTED]Schuh, Richard [EMAIL PROTECTED] Schuh, Richard [EMAIL PROTECTED] Sent by: The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU 11/27/2007 01:55 PM Please respond to The IBM z/VM Operating System IBMVM@LISTSERV.UARK.EDU To IBMVM@LISTSERV.UARK.EDU cc Subject CRR Machine Can the CRR machine be taken down for s brief interval without taking down other SFS servers, or is it necessary to stop all SFS activity to do it? The situation is that the disks of the CRR machine are on physical disks that are to be decommissioned and I need to move them to a more permanent location. Naturally, if this were possible, it would take place during a period of extremely low activity. Regards, Richard Schuh
Re: CRR Machine
On Tue, 27 Nov 2007 13:27:08 -0800, Schuh, Richard [EMAIL PROTECTED] wrot e: Are the warnings about Limp Mode if there is no CRR machine still extant? If so, there is still something being done by the CRR machine. When lightly loaded, I would not expect it to be a problem. Regards, Richard Schuh DB2 definitely uses CRR, and possibly other applications. There are CRR c ommands that might tell you something, although I have not tried them. Limp Mode is still documented in z/VM V5R3.0 Performance. If the CRR server is not running, users who use SFS (and other resources that participate in CRR) are in a condition called limp mode. While in this condition, two-phase c ommits are not possible. Therefore, applications that use protected conversations or other resourc es that do not support simple commit logic cannot be run. A user may experience significant SFS performance degradation while in limp mode. To avoid limp mode, IBM strongly recommen ds that every such system have the CRR server running. The generation of a CRR server (VMSER VR) and its associated file pool (VMSYSR) are optional z/VM installation tasks. The question is, when are two-phase commits used? From what others have t old you, one answer is when you have R/W access to two different SFS filepools. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
Re: CRR Machine
A VM DB2 server can indeed use CRR, but, (as far I know) not for DB2 applications running on VM. That is, DB2/VM uses CRR only when it used in DRDA mode from a remote site. 2007/11/28, Alan Ackerman [EMAIL PROTECTED]: On Tue, 27 Nov 2007 13:27:08 -0800, Schuh, Richard [EMAIL PROTECTED] wrot e: Are the warnings about Limp Mode if there is no CRR machine still extant? If so, there is still something being done by the CRR machine. When lightly loaded, I would not expect it to be a problem. Regards, Richard Schuh DB2 definitely uses CRR, and possibly other applications. There are CRR c ommands that might tell you something, although I have not tried them. Limp Mode is still documented in z/VM V5R3.0 Performance. If the CRR server is not running, users who use SFS (and other resources that participate in CRR) are in a condition called limp mode. While in this condition, two-phase c ommits are not possible. Therefore, applications that use protected conversations or other resourc es that do not support simple commit logic cannot be run. A user may experience significant SFS performance degradation while in limp mode. To avoid limp mode, IBM strongly recommen ds that every such system have the CRR server running. The generation of a CRR server (VMSER VR) and its associated file pool (VMSYSR) are optional z/VM installation tasks. The question is, when are two-phase commits used? From what others have t old you, one answer is when you have R/W access to two different SFS filepools. Alan Ackerman Alan (dot) Ackerman (at) Bank of America (dot) com
