Hi,
We are planning to migrate existing UNIX AIX 5.3 Server to new UNIX AIX 5.3 Server.. Same hardware configuration is used in the new server ie.,as same as old server..
Could you please advise me.. for new scenario which one is best option..
UNIX Cloning -- According to me its the fastest and safest.
[By doing this no need to install datastage and its components].
My question is if we clone "Do we need to compile the existing job's again?". And also is there any "disadvantes using cloning for Datastage?"
or
Fresh Installation of O/S and Datastage
Thanks in advance...
Can we do Datastage Cloning?
Moderators: chulett, rschirm, roy
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
You may get "duplicate hostname on network" errors.
DataStage might work on the independent machines, or it might not because of this. It is required that Information Server components be on a machine that can resolve its hostname to a unique IP address.
DataStage might work on the independent machines, or it might not because of this. It is required that Information Server components be on a machine that can resolve its hostname to a unique IP address.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
How many tiers do you have?
I would advise against just a disk cloning. You'll have hostname issues in the xmeta. You'll have port issues.
If it were me, I'd keep the existing server up and running right up until you'll dumpped all of the projects/jobs/scripts/parms/kernel settings/ssh keys/etc... to the new box.
Patch it up to the latest and greatest code release level (now that you have a chance to take that server outage window). Export your jobs into an ISX file via the IS Manager tool, and re-import into location #2. Set a "freeze date" for your customers where the projects will cut over. If it's the same release level of DataStage, then you might not need a recompile. But as an admin, you would want to recompile some of your testcase jobs before porting over the whole lot. You don't want egg on your face if it fails.
I would advise against just a disk cloning. You'll have hostname issues in the xmeta. You'll have port issues.
If it were me, I'd keep the existing server up and running right up until you'll dumpped all of the projects/jobs/scripts/parms/kernel settings/ssh keys/etc... to the new box.
Patch it up to the latest and greatest code release level (now that you have a chance to take that server outage window). Export your jobs into an ISX file via the IS Manager tool, and re-import into location #2. Set a "freeze date" for your customers where the projects will cut over. If it's the same release level of DataStage, then you might not need a recompile. But as an admin, you would want to recompile some of your testcase jobs before porting over the whole lot. You don't want egg on your face if it fails.
Single Tier for all server components:PaulVL wrote:How many tiers do you have?
I would advise against just a disk cloning. You'll have hostname issues in the xmeta. You'll have port issues.
(Engine Tier, Services Tier, Metadata Repository Tier) installed in single pc.
we set-up new sever and both old and new servers are look's fine all files/folders copied with appropritae permissions..
I am trying to modify XMETA hostname for new server:
/opt/IBM/db2/V9.1 --> default.env
Could you please guide me where I need to modify the host name & port number.. or any other configurations required..
Thanks,
Sekhar
Sekhar