RE: PowerHA for i questions -- MIDRANGE-L

Evan,

You don't have to start out with an empty iASP on the source, although if it
is empty the initial synch will be very fast. The target iASP is built when
you create the clusters and other objects for PowerHA. I'm not sure what
you're referring to when it says you can't add an iASP that exists to a
cluster. If that is the target iASP then yes that's true, but the source
can, and most likely will be there.

Yes the "Heartbeat" and "Replication" LANs need to be on different VLANS and
as isolated as practical. If the "Heartbeat" network (administrative)
loses communications for a short time PowerHA thinks there might be a
failure. The replication traffic needs to be isolated as well and should be
on a fairly beefy communications line. The cluster does not have to manage
the NICs but certainly there's no reason not to have them part of the
administrative domain.

The two problems I run into with PowerHA are first the communications line
between two locations does not have sufficient capacity and secondly not
enough CPU is allocated to PowerHA. If either of those two gets saturated
then the cluster starts to fall behind and will suspend to protect the
source system.

The initial synch happens when you create the target iASP using the PowerHA
commands to start the replication (I don't remember the exact command and
I'm not near a machine)

--
Jim Oberholtzer
Chief Technical Architect
Agile Technology Architects

-----Original Message-----
From: MIDRANGE-L [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Evan
Harris
Sent: Wednesday, May 06, 2015 6:45 PM
To: Midrange Systems Technical Discussion
Subject: Re: PowerHA for i questions

Hi Patrick,

thanks for that, that's really useful. As far as building the IASP part of
the equation, I think the light bulb just went on. I need to start with an
empty IASP - which is why the book says you can't add IASPs that already
exist into a cluster.

It's the address the users connect with that is exercising my thinking at
the moment.

You say that "if on the same layer 2 VLAN, you'll want an IP that your users
will connect with" so presumably the NICs are resources the cluster manages.
I could not find any specific pages in the books describing this but I'll go
back and look.

And if they are on different networks ? I suppose you just modify your DNS
and let nature take it's course to get up and running again.

On Thu, May 7, 2015 at 11:18 AM, Patrick.Bingham@xxxxxxxxxxxxx <
Patrick.Bingham@xxxxxxxxxxxxx> wrote:

1) You'll want an administrative IP address for each system base. And
then at least one IP address for each system base that cluster
services will use to communicate. You could share the administrative
IP but I wouldn't call that "Best Practices". Then, if on the same
layer 2 VLAN, you'll want an IP that your users will connect with. It
will be a cluster resource that can be active when the iASP is active.

2) The synch will be done when you restore data into the source system
iASP. Cluster replication (or storage replication if you're using it)
will synchronize the blocks. Anything in the target iASP will be
written over when cluster is setup. There is no "priming the pump".

3) The system name can be the same. Cluster services uses the node name.
The RDB name will be assigned to the iASP and follow where the iASP is
running.

Hopefully that's a good start! Good luck

Patrick Bingham | Principal, Converged Infrastructure Architect
Office: 402-965-2381 | Mobile: 402-212-2944
patrick.bingham@xxxxxxxxxxxxx Sirius Computer Solutions |
www.siriuscom.com
14301 FNB Parkway, Suite 400, Omaha, NE 68154

-----Original Message-----
From: Evan Harris [mailto:auctionitis@xxxxxxxxx]
Sent: Wednesday, May 06, 2015 5:38 PM
To: Midrange Systems Technical Discussion
Subject: PowerHA for i questions

Hi All

I have been doing some research into PowerHA for a possible Proof of
Concept implementation. Most of my reading has been in the
"Implementing PowerHA for IBMi" and PowerHA SystemMirror for IBM i

Cookbook" redbooks.

The scenario I am looking at initially is using two guest LPARs on two
Power 7 machines as a proof of concept. I know how to build a guest
LPAR and also have the IASP side of things under control but I do have
a couple of questions that I have not been able to find in the reference

material.

1. What are the network requirements (if any) and how do I set up the
network for failover

Do I assign one address that the cluster group then uses to direct
traffic to the live node, or do I do something like change the DNS
entry to point the host name to a different IP address ? Or some other

approach ?

2. How do I initially synch up the IASP's

Do I have to do a save and restore of the IASP or the library
contents, or do I need to let the Cluster perform a some kind of synch
operation ? Will restoring the data into the target IASP make this any

quicker ?

3. Can I have the same system name (SNA system name) on both nodes in
the cluster ?

I ask this last question simply out of curiousity as a customer I am
dealing with has software that they insist must have a specific system
name to function. I don't think this is ideal, nor do I think it will
work but I can't think of a specific reason why it won't. I suspect
the cluster group won;t allow to nodes to have the same system name
but haven;t been able to verify this so far.

If anyone can give me some pointers on these questions or point me at
some other reference material that would be great.

--

Regards
Evan Harris
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe,
unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx Before posting, please take
a moment to review the archives at http://archive.midrange.com/midrange-l.

This message (including any attachments) is intended only for the use
of the individual or entity to which it is addressed and may contain
information that is non-public, proprietary, privileged, confidential,
and exempt from disclosure under applicable law. If you are not the
intended recipient, you are hereby notified that any use,
dissemination, distribution, or copying of this communication is
strictly prohibited. This message may be viewed by parties at Sirius
Computer Solutions other than those named in the message header. This
message does not contain an official representation of Sirius Computer
Solutions. If you have received this communication in error, notify
Sirius Computer Solutions immediately and (i) destroy this message if
a facsimile or (ii) delete this message immediately if this is an

electronic communication. Thank you.

Sirius Computer Solutions<http://www.siriuscom.com>
--
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe,
unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx Before posting, please take
a moment to review the archives at
http://archive.midrange.com/midrange-l.