To expand just a little on what Adam just said.....

Adam Thornton wrote:

On Mon, 2004-05-10 at 09:16, Phil Hodgson wrote:


Does anyone run any hot standby systems on Linux 390?
Is the data (if any) migrated and health checked by your own code or is there some 
software that I need to be made aware of?



There are certainly sites using the Linux-HA clustering code on 390. Although that's usually active-active rather than hot-standby.



Both systems are up and running the same apps at the same time. When one
fails,  the routers redirect incoming traffic to the second system, and
it can then access the DASD of the failed first system. Works neat.....

You can also certainly define a second set of machines on a second frame
with access to the same DASD, as a full-frame (or, with appropriate use
of VIPA, a per-image) failover configuration.  To do this really really
right and with neat automation, you could use CSE, but even without it
it's doable as long as your staff is trained to know that you never ever
ever run the same image on both frames simultaneously.  This gives you
nearly-hot standby: when one image fails, a monitor of some kind
(doesn't have to be sophisticated) kicks off its opposite number on the
other frame.  Failure detection takes some amount of time, and then IPL
of the new guest takes a little longer.



The CSE that Adam mentions is the Cross Systems Environment  (CSE) for
z/VM.  With it, two or more VM systems can share 1) the CP directory, 2)
spool space,  and 3)  minidisks in a controlled and safe manner. Of
course, this approach to Linux high availability systems only works with
Linux running as guests of the VM systems.

Feel free to contact me off-list if you'd like all the gory details
about setting up and using CSE in a HA Linux environment. The nice thing
about CSE is that, if you have z/VM now, you already have CSE as well.
It's just a matter of configuring and starting it up......

You could combine an approach like this with FlashCopy or PPRC or both,
if you had the disk space to spare.  Then you could recover the original
image in parallel while its spare was running.

Adam

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




-- Dave Jones Sine Nomine Associates Houston, TX 281.578.7544

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

Reply via email to