Hall, Ken (GTS) wrote:
Even if you "preformat" the DASD, it still has to be done, so it does
take time.  And if you're using minidisks, you have to format it at the
point where the guest is created.

On a related note, anyone know how to eliminate the prompt during
kickstart?:

Can't have a question in command line mode!

Warning

The partition table on device dasdb (0.0.0101) was unreadable. To create
new par
titions it must be initialized, causing the loss of ALL DATA on this
drive.


This operation will override any previous installation choices about
which drive
s to ignore.



Would you like to initialize this drive, erasing ALL DATA?

yesno []



I tried to format over a partially formatted minidisk, and got this.
There doesn't seem to be an option to override and just go ahead and
write.

This is RHEL5.3.


The combination of kickstart options:

clearpart --all --initlabel
zerombr yes

will instruct the installer to go ahead and format the DASD without
prompting.

To your first point, I think Doug was pointing out that you don't have
to do this format every time.  If you have a DASD that was formatted for
Linux use in the past, you can just specify:

clearpart --all

and Anaconda will happily skip the dasdfmt.  This works great for DASD
used by Linux in the past.  For unknown or new DASD, it's best to use
the slow approach (dasdfmt during install), that way you know for sure.

-Brad

-----Original Message-----
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of
William D Carroll
Sent: Wednesday, June 10, 2009 12:57 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] To kick or to clone ... that is the question

As I have said though
kickstart is faster if you do not need to format.
In our case we preformat dasd when we get it so this is not needed.
a point that seems to keep getting over looked.

and point out your test becomes 5 min for kickstart and 16 min for clone
if dasd is performated.


William 'Doug' Carroll
Mainframe Systems Eng Sr I
Global Technology Infrastructure


-----Original Message-----
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of
Hall, Ken (GTS)
Sent: Wednesday, June 10, 2009 12:51 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: To kick or to clone ... that is the question

Okay, I gave up and ran timings.

Our base system consists of 7 minidisks, for a total of 22,744
cylinders.  About half of that (three volumes) is empty space in an LV
for a product that generates data after the system is up.  The actual
base system is three minidisks, plus one swap disk, that roughly fits on
a 3390-9.  It all has to be either formatted or copied.

Kickstart took 21 minutes to format all that, plus an additional 5
minutes (surprisingly) to install 491 packages, for a total of 26
minutes.  The process failed in the end because of a bug in my
post-install script, but the system was manually bootable.

Our clone process took 16 minutes to sequentially copy the same 7 disks
via DDR.  If you're clever, you can set up parallel service machines to
perform the copies at once and get it down to maybe 5 minutes total
(assuming your system isn't all on one disk).  We had done this in a
previous incarnation of the Linux project, but our current method is
sequential.

As I said, the difference is significant in itself, but we're talking
minutes.  Different combinations of disks with different numbers of
packages, and amounts of empty space will affect the results, so Your
Mileage May DEFINITELY Vary.

My main gripe with kickstart is that it's complicated, buggy (on z), and
badly documented (on z).  I spent a lot of time fooling with it during
our original setup (gave up), once again later (when I was finally able
to get it to work), and because we may have to switch to using it, I'm
still trying to get it working properly in my "spare" time.



-----Original Message-----
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of
William D Carroll
Sent: Wednesday, June 10, 2009 11:04 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] To kick or to clone ... that is the question

I cannot agree with your statement
"Any way you cut it, there's still more overhead in kickstart, and it
still takes longer"
as I've stated unless you're using flashcopy (not available to everyone)
a DDR can take longer than a Kickstart, Real experience here with
kickstarting being faster than DDR's
kickstarting can be real simple as well.  php to create the kickstart
pull IP and info from a DB/flatfile.  server kickstarted and up in <5min
(if no dasdfmt needed)

I'm curious, how many have actually tried kickstarting servers and
really dived into it?
not just a sample kickstart and then didn't' like it.  really looked at
it?


William 'Doug' Carroll
Mainframe Systems Eng Sr I
Global Technology Infrastructure
RedHat Certified Engineer:  805008304430937


-----Original Message-----
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of
Hall, Ken (GTS)
Sent: Wednesday, June 10, 2009 10:31 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: To kick or to clone ... that is the question

Well, actually, all our guests use hipersockets, for reasons I won't get
into, but we also use vswitch, and ping times going between two guests
on a vswitch are comparable to guests on the hipersocket.

Any way you cut it, there's still more overhead in kickstart, and it
still takes longer.  Even cutting out the formatting time, you still
have to read the packages, send them over the network, and install them,
but in the end you're talking minutes.  There are other advantages to
kickstart, I just wish it worked better and was better documented for
zSeries.  I had a terrible time figuring out how to get it working the
first time, and a few of the parameters don't seem to work as advertised
(or at least didn't with RHEL5.2).

If you could put the kickstart file on the 191 disk, IPL the kernel,
have it install headless, and reboot properly afterward, it would be
worthwhile, but I could never get it to do that, and the kickstart file
has to be on a server somewhere.  At the moment, we control everything
from VM, and creating instances is as simple as it can get.


-----Original Message-----
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of
Mark Pace
Sent: Wednesday, June 10, 2009 10:12 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] To kick or to clone ... that is the question

Unless your installation source is on a Hipersocket connection, then the
network time is near zilch.

On Wed, Jun 10, 2009 at 10:02 AM, Hall, Ken (GTS) <ken_h...@ml.com>
wrote:

The small addition of the read time for DDR is nothing compared to the
time required to install the packages over the network.  That's at
least
the same amount of reading, plus network time.

I agree about the problem of keeping multiple masters.  We keep just
one
per version, and use a first-boot process to install additional
packages
via yum as needed, or let the admins install what they need manually.

It's still faster, by quite a lot.  Unfortunately, I don't have time
today to run measurements, but our base system clone process takes
less
than 30 minutes from beginning to end.

All this said, we have been seriously considering going to a kickstart
based method, but my experience with it has not been encouraging.
Aside
from taking longer, it seems to be fairly fragile and requires more
manual effort.  Our clone method consists of running a VM-based
dialog,
waiting for the copies to finish (run asynchronously in a service
machine), and then autologging the new guest.


-----Original Message-----
From: Linux on 390 Port [mailto:linux-...@vm.marist.edu] On Behalf Of
Rob van der Heij
Sent: Wednesday, June 10, 2009 9:16 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] To kick or to clone ... that is the question

On Wed, Jun 10, 2009 at 2:54 PM, Hall, Ken (GTS)<ken_h...@ml.com>
wrote:
How could it be faster?

Cloning involves simply copying the disks, that's one pass with DDR.
Copying a disk requires reading and writing. Formatting just requires
(short) writes. Depending on your configuration, you may not notice
the extra resource usage in the elapsed time.

But it's probably more whether you spend the time while you're waiting
for it. Once you get into the business of holding several different
golden masters to copy from, things get more complicated.

Back then we used a very bare minimum that was copied to the new root
disk, and the required additional packages were added on top of it.
That approach allows for a stock supply of copied root disks ready to
use.

-Rob

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390
or
visit
http://www.marist.edu/htbin/wlvindex?LINUX-390


------------------------------------------------------------------------
--
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. References to "Merrill Lynch" are references to any
company in
the Merrill Lynch & Co., Inc. group of companies, which are
wholly-owned by
Bank of America Corporation. Securities and Insurance Products: * Are
Not
FDIC Insured * Are Not Bank Guaranteed * May Lose Value * Are Not a
Bank
Deposit * Are Not a Condition to Any Banking Service or Activity * Are
Not
Insured by Any Federal Government Agency. Attachments that are part of
this
E-communication may have additional important disclosures and
disclaimers,
which you should read. 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.

------------------------------------------------------------------------
--

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390
or
visit
http://www.marist.edu/htbin/wlvindex?LINUX-390




--
Mark Pace
Mainline Information Systems
1700 Summit Lake Drive
Tallahassee, FL. 32317

http://www.marist.edu/htbin/wlvindex?LINUX-390

--------------------------------------------------------------------------
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. References to "Merrill Lynch" are references to any company in the 
Merrill Lynch & Co., Inc. group of companies, which are wholly-owned by Bank of America 
Corporation. Sec


urities and Insurance Products: * Are Not FDIC Insured * Are Not Bank 
Guaranteed * May Lose Value * Are Not a Bank Deposit * Are Not a Condition to 
Any Banking Service or Activity * Are Not Insured by Any Federal Government 
Agency. Attachments that are part of this E-communication may have additional 
important disclosures and disclaimers, which you should read. 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.
--------------------------------------------------------------------------


----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

--
Brad Hinson <bhin...@redhat.com>
Sr. Support Engineer Lead, System z
Red Hat, Inc.
(919) 754-4198
www.redhat.com/z

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to lists...@vm.marist.edu with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390

Reply via email to