Hi Tom,
Thanks for all the information. I really do appreciate it.
Yeah, I understand about power outages. We had one here a couple years ago.
One UPS failed out right (37 dead cells), and the second one failed due to
human error ("Don't touch that switch!". "You mean this switch?".). My 9672
and Disk Arrays stayed up just fine (internal batteries). I did lose Tape and
Console controllers. 3 Test/Dev Systems stayed up, and once the Console
Controllers were back they reconnected just fine. Production would have stayed
up, if the Operator hadn't followed IBM Remote Support's directions to try and
IPL to determine why the Console wasn't responding (remember those dead
controllers?).
Thanks,
Dave
Dave Stuart
Prin. Info. Systems Support Analyst
County of Ventura, CA
805-662-6731
[EMAIL PROTECTED]
>>> "Tom Duerbusch" <[EMAIL PROTECTED]> 8/15/2007 11:43 AM >>>
Long story.
The City of St. Louis succeeded from St. Louis County back in the late 1880's.
So the City of St. Louis is also the County of the City of St. Louis (not to be
confused with St. Louis County). As such City Hall contains the Mayor of the
City plus 9 County Officers (some elected, some appointed by the State of
Missouri).
So there are 10 chiefs and no single person in charge.
I.S. (A City department) is suppose to cover the City, and any needs the County
offices has (if they are willing to kick in some bucks).
The CIO, sees great need and great promise of server consolidation. Just no
money. Trying to get the "county" offices onboard, slows up when the offices
need to kick in some cash, until they understand the cost savings.
In the last hardware upgrade we bought an IBM z/890 130 with an IFL and went
from 400 GB dasd to 4 TB dasd. That all happened two years ago.
We still are not in production with much on the IFL side. Lots of proof of
concepts, and some development.
What we have and have tried are:
Oracle 10g(s): Will go into production. Many departmental databases will be
migrated there.
DB2/UDB(s): Will go into production. The DB2/VSE engine will be moved here.
Some other departmental databases will also be migrated here.
SAMBA Server(s): Minor use in production. Replacement for LANRES/VSE Disk
Host function.
FTP Server(s): Minor use in production. Replacement for LANRES/VSE
Distribution function. Also used for the install server for zLinux.
NFS Server(s): File server for Linux applications.
VSE VTAPE Server(s):
GPG Server(s): VSE files are sent here to be encrypted prior to be sent out on
the public Internet.
Apache Web Server(s): If mainframe data is being served, it will be served by
a mainframe web server, over hipersockets.
Future (within 12 months)
Websphere/HATS: We will be testing it as a front end to webafry our home grown
applications (also z/WebHost on the VSE side will be tested)
Websphere/HOD: Host on Demand. This is a purchased part of our GEAC
accounting system called "active client". It webafry the accounting system.
Keymanager for 3592 encryption tapes: When we get the tape drives purchased.
SSLServer: For TN3270 sessions.
PDF Server: There is talk about taking certain mainframe reports (W2, 1099,
tax bills) and storing them in PDF format, so the end users can reprint a
single page when needed.
Right now, between test, development, proof of concept and my "testing"
servers, we have 50 (Linux50 is my first attempt at SLES10) servers created
with about 2 dozen running.
My push is that if it can run on Linux, and there is a z distribution for it,
try it on the mainframe. Zero initial cost (everything is already bought),
I'll give you a server in a couple hours (instead of a couple months to buy
one) and if the application goes into production, we can make a decision on if
it should stay on the mainframe, or offloaded to another box. I expect most
stuff to stay.
Current IFL CPU utilization over 1st shift, about 15%. Like I said, very few
users.
Last week, our UPS failed and shutdown power to the entire machine room.
Funny, the IBM z/890 kept running. It took 10 minutes of futile frustration as
no I/O devices were available. However, even though the IBM DS6800 drives spun
down, the controller and cache stayed up and recorded the "attempts" at error
recovery from the z/890. Eventually, I shutdown the mainframe (didn't need to)
as we switch from the UPS to the electric company.
We would have stayed up. No UPS, no electric power. At least long enough to
bring back power.
How is that for uptime?
Tom Duerbusch
THD Consulting
(sometimes I wonder, What am I doing here?)
>>> David Stuart <[EMAIL PROTECTED]> 8/14/2007 6:12 PM >>>
Afternoon,
We are looking at bringing in a z9 BC to replace our existing 9672, with the
possibility of also running Linux on the z9. However, my CIO is concerned with
the types of applications, number of IFLs required, other county governments
doing this, etc., that can be run on the z9.
Are there any county governments running Linux on a z-Series or z9?
If so, would you be kind enough to share the types of applications, number of
IFL's, data bases, or any other information that might be of interest to my
CIO.
TIA,
Dave
Dave Stuart
Prin. Info. Systems Support Analyst
County of Ventura, CA
805-662-6731
[EMAIL PROTECTED]
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
