I will address how I have implemented #1 - end user HOME subdirectories. I use 
automount. I do this as follows:

In /etc/rc (UNIX startup script), I put the lines:

# Start the automount facility
/usr/sbin/automount


I created the UNIX file /etc/automaster containing the line:

/home /etc/home.map

I use "/home" as the "home" root subdirectory. Many others use "/u". Modify the 
above as necessary. In /etc/home.map, I have:

name *
type ZFS
Filesystem TSSPV.PLEX&SYSPLEX..HOME.<uc_name>.ZFS
mode rdwr
allocuser space(20,10) cyl dataclas(DCSTD)
parm FSFULL(70,10)
delay 10
duration 10
setuid No

I don't need to do much of anything after this. When a user logs into a UNIX 
shell for the first time, a VSAM zFS filesystem is created and initialized, 
then mounted. Thereafter, the existing zFS data set is mounted. The "<uc_name>" 
is replaced by the upper case value of the subdirectory. In the OMVS segment, I 
have something like:

OMVS INFORMATION                                                                
----------------                                                                
UID= 0000001009                                                                 
HOME= /home/tsh009                                                              
PROGRAM= /bin/sh                                                                
CPUTIMEMAX= NONE                                                                
ASSIZEMAX= NONE                                                                 
FILEPROCMAX= NONE                                                               
PROCUSERMAX= NONE                                                               
THREADSMAX= NONE                                                                
MMAPAREAMAX= NONE         


Unfortunately, this does not lead to easy use of the automatic assignment of an 
OMVS segment because I end up "hard coding" the HOME and it cannot be coded 
with some sort of variable. The "<uc_name>" previously mentioned comes from the 
"/home/tsh009" in the HOME field. Automount strips out the "/home/" portion, 
then upper cases the remaining "tsh009", so the zFS data set name is 
TSSPV.PLEXLIH1.HOME.TSH009.ZFS . Note that I, perhaps unnecessarily, include 
the PLEXLIH1 node. I'm not sure why I really do this. Perhaps due to the 
"strange" way that I assign temporary space to a UNIX shell user. I can go into 
that if you like, but basically I use automount to dynamically create a system 
specific zFS filesystem for each user and "hard code" them to more-or-less 
force each user to use their own zFS "temporary" file system, instead of 
generally using the "/tmp" mount point. I do this so that one "greedy" user 
cannot use up all the UNIX temporary space and thus cause a problem to other 
UNIX users.


-- 
John McKown
Systems Engineer IV
IT

Administrative Services Group

HealthMarkets(r)

9151 Boulevard 26 * N. Richland Hills * TX 76010
(817) 255-3225 phone *
[email protected] * 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

> -----Original Message-----
> From: IBM Mainframe Discussion List [mailto:[email protected]]
> On Behalf Of Jousma, David
> Sent: Wednesday, October 31, 2012 9:48 AM
> To: [email protected]
> Subject: JAVA development best practices
> 
> All,
> 
> Looking for some feedback - publications, real world experience, etc on
> managing the z/OS environment to support JAVA application development
> in our shop.   Up until now, JAVA has been a service provider for
> systems level(and controlled) functions.
> 
> A couple things come to mind immediately that have never been a real
> concern before, and maybe are a pipe dream:
> 
> 
> 1)      Unix Systems Services for the masses - today no one outside of
> tech services can get to unix command prompt.   I suspect developers
> will need their own filesystems, etc.  Don't want to make it
> complicated or require a lot of administrative overhead.  Security?
> 
> 2)      Isolating JAVA release levels from applications.  Is this
> possible?   We tend to keep several versions of JAVA available(1.4,
> 5,6, and 7) but make the most current, the "default" as in the path in
> JAVA_HOME.   What I don't want is applications coding for specific
> versions(if possible) causing upgrade nightmares.
> 
> 3)      Shared file systems.  Today we do not share filesystems for
> read/write across the lpars in our sysplex.   We share sysres
> filesystems readonly.  Filesystems for write access are system specific
> today.
> 
> 4)      ??
> 
> What am I missing?   Please share your experiences either on/off list
> is fine.   Just looking to learn from others and be able to implement
> general best practices BEFORE the masses hit the systems.
> 
> Thanks, Dave
> 
> _________________________________________________________________
> Dave Jousma
> Assistant Vice President, Mainframe Services [email protected] 1830
> East Paris, Grand Rapids, MI  49546 MD RSCB2H p 616.653.8429 f
> 616.653.2717
> 
> This e-mail transmission contains information that is confidential and
> may be privileged.
> It is intended only for the addressee(s) named above. If you receive
> this e-mail in error, please do not read, copy or disseminate it in any
> manner.  If you are not the intended recipient, any disclosure,
> copying, distribution or use of the contents of this information is
> prohibited. Please reply to the message immediately by informing the
> sender that the message was misdirected. After replying, please erase
> it from your computer system. Your assistance in correcting this error
> is appreciated.
> 
> 
> 
> 
> ----------------------------------------------------------------------
> 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

Reply via email to