This does beg the question, is there any significant performance
difference when writing to the filemode vs. writing to an unaccessed
directory?
No, when the destination for the control data backup is an SFS directory,
the same routines are used, regardless of whether the directory is
Of Sue Farrell
Sent: Friday, July 25, 2008 11:05 AM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: Size of SFS control backup
Speaking of Chapter 7, this is a must read for all
responsible for maintaining SFS file pools. The control data
backup that we have been =
talking about is only HALF
I'm a little late to this discussion, but there are a couple things I
wanted to respond to:
Thanks! That explains why I filled it up and how big it should be. Did
I miss it, or is it really not in book?
-Rob
There is a section titled DASD Space Needed for Control Data Backup in
Chapter 7
On Tuesday, 07/22/2008 at 02:30 EDT, Rob van der Heij [EMAIL PROTECTED]
wrote:
On Tue, Jul 22, 2008 at 8:28 AM, Rob van der Heij [EMAIL PROTECTED]
wrote:
The description of this process does not look like it was for
production use.
Maybe we need SFS to be enhanced to access the filemode
On Wed, Jul 23, 2008 at 8:25 PM, Sue Farrell [EMAIL PROTECTED] wrote:
You can direct the control data backup to another file pool using the
explicit directory name and thus, avoid the access, and the problem of the
access going away.
Eg. fileserv defbackup disk backup data fpname:username.dir
On Wed, Jul 23, 2008 at 8:31 PM, Alan Altmark [EMAIL PROTECTED] wrote:
Use an ENDCMD nucleus extension to check for and re-access any missing
directories.
You mean in the SFS server to catch it when it falls in CMS Ready ? Yuck.
And it's way more useful than just there. When we ran CMS
Sent: Wednesday, July 23, 2008 3:12 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: Size of SFS control backup
On Wed, Jul 23, 2008 at 8:31 PM, Alan Altmark
[EMAIL PROTECTED] wrote:
Use an ENDCMD nucleus extension to check for and re-access
any missing
directories.
You mean
On Tue, Jul 22, 2008 at 6:48 AM, Kris Buelens [EMAIL PROTECTED] wrote:
No need to restart SFS. One should learn about the .CMS SFS operator
command. The problem is that one first neads to zap a bit in the
SFS server's storage before .CMS is accepted. Look for and SFSDOT
file on one of
On Tue, Jul 22, 2008 at 8:28 AM, Rob van der Heij [EMAIL PROTECTED] wrote:
The description of this process does not look like it was for production use.
Maybe we need SFS to be enhanced to access the filemode as specified
before trying to backup. Or my long-standing desire for CMS itself to
I was rather surprised to see my disk for the control backup of the
filepool fill up.
I understand the backup file can only be as large as half of the disk.
But my log disks are 20 cyls, and the backup disk is 100 cyls. So that
should fit, no?
-Rob
.
Regards,
Richard Schuh
-Original Message-
From: The IBM z/VM Operating System
[mailto:[EMAIL PROTECTED] On Behalf Of Rob van der Heij
Sent: Monday, July 21, 2008 1:03 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Size of SFS control backup
I was rather surprised to see my disk
.
Regards,
Richard Schuh
-Original Message-
From: The IBM z/VM Operating System
[mailto:[EMAIL PROTECTED] On Behalf Of Rob van der Heij
Sent: Monday, July 21, 2008 1:03 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Size of SFS control backup
I was rather surprised to see my disk
On Mon, Jul 21, 2008 at 10:22 PM, Kris Buelens [EMAIL PROTECTED] wrote:
The control data backup is a backup of storage group 1, plus the
control minidisk, and the POOLDEF file. Its size is not related at
all to the size of the log files. The log file size merely defines
how often a control
I'd say it is in the book. We used to have an automatic, daily,
check. You only need the space of what is actually used in storage
group 1, i.e. not how big storage group 1 is defined.
2008/7/21 Rob van der Heij [EMAIL PROTECTED]:
On Mon, Jul 21, 2008 at 10:22 PM, Kris Buelens [EMAIL
On Mon, Jul 21, 2008 at 10:59 PM, Kris Buelens [EMAIL PROTECTED] wrote:
I'd say it is in the book. We used to have an automatic, daily,
check. You only need the space of what is actually used in storage
group 1, i.e. not how big storage group 1 is defined.
Yeah, until someone does a lot of
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: Size of SFS control backup
On Mon, Jul 21, 2008 at 10:59 PM, Kris Buelens
[EMAIL PROTECTED] wrote:
I'd say it is in the book. We used to have an automatic, daily,
check. You only need the space of what is actually used in storage
group 1, i.e
On Mon, Jul 21, 2008 at 10:16 PM, Schuh, Richard [EMAIL PROTECTED] wrote:
affects ever user that is actively using SFS. The documentation also
says that backing up to a different filepool instead of to dasd or tape
will cause less interference with the users. It will take longer, but
the
@LISTSERV.UARK.EDU
Subject: Re: Size of SFS control backup
On Mon, Jul 21, 2008 at 10:16 PM, Schuh, Richard
[EMAIL PROTECTED] wrote:
affects ever user that is actively using SFS. The
documentation also
says that backing up to a different filepool instead of to dasd or
tape will cause less
18 matches
Mail list logo
