Hi everyone,

We have recently implemented SAN replication for high availability on our IBM iSeries environment. The replication is set up between our production SAN Storwize V5000E Storage Management (DTV5030E) and DR environments using IBM FlashSystem 7300 (AS400FS7300-1). The data synchronization is in progress, and once the replication is complete we need to power on the DR machine and boot it from the SAN replicated LUNs.

Can you please help to check the completion status of the replication and the LUNs that are to be noted for powering the DR machine using the SAN replicated data.

Could anyone provide a step-by-step guide or point me to the relevant IBM documentation that outlines the procedure for configuring the DR LPAR to boot from the SAN replication LUNs? Specifically, I'm looking for details on:

• LPAR naming considerations (same or unique names for DR).
• Adjustments needed in the HMC console.
• Any network or DNS changes required for the DR environment.
• Best practices for ensuring the DR machine starts correctly from the SAN LUNs.

Any insights, documentation links, or screenshots would be greatly appreciated!

Thanks in advance for your help!

Dear Suresh

I think that checking the completion of the data replication must be done only from the source storage device and you stand a better chance asking this in Storage group. 

As for the first 3 points you asked, I do not see there is any strict requirements to follow. Perhaps, using common sense is the way. 

For the last point on how to ensure the DR machine starts correctly from the SAN LUNs, you need to put the Load Source Tagging (from HMC) to the specific disk controller cards (or vFibre Card) that connects to the specific LUN that is the Load Source Disk.  (There is no issue if there is only one fibre card used to connect to all those LUNs.)   Which LUN in the target SAN disk is the Load Source, you also need to ask in Storage group. 

Also check this redbook in case it has useful info for you: IBM i and IBM Storwize Family- A Practical Guide to Usage Scenarios: https://www.redbooks.ibm.com/redbooks/pdfs/sg248197.pdf

Dear Suresh,

we are working a lot with storage replication for IBM i LPARs. It is a very reliable solution, and if the replication is setup correctly using consistency groups, there is no need for a day to day check if everything is sync between the storages. 

Regarding your questions it is hard to deliver final answers / recommendation without knowing your setup. E.g. the distance between the locations, or if you are using FC or IP replication is having some impact. 

We are using different LPAR names on the different locations. This helps us to identify if a partition is currently running on the primary or secondary location (eg. LPAR01 for prod, LPAR02 for DR, LPAR03 for Test ...). There are no special requirements for adjustments on the HMC or console for such a setup.

Regarding network it is very helpful if you can use a Layer 2 network between the Prod and the DR location, meaning that you can use the same IP Address on Prod and DR location. If this is in place there is no need for any system adjustments in the case of a failover, you just need to start the LPAR.

If you are using 3rd party license program that are linked to a serial number or LPAR ID, it might be necessary to apply keys if you switch between the locations. If you have such license programs you can write a simple CL program that is applying the keys during the IPL based on the serial number the LPAR is starting on. All IBM i license keys you can apply for Prod and DR system upfront.

Regarding SAN it is important that the zoning is in place on the DR location, similar to the production location. E.g. if you are using NPIV you need to create all the virtual adapters and do the necessary config. 

If you are using a Fiber Channel attached tape drive you might also want to have access to the tape drive independent if the system is running on location A or B. 

I hope this helps.

Klaus