HPE MSA 2052 remote replication

[AllanClark]

Hello,

I am a newbie to replication technolgies with HPE MSA 2052s. We are replacing 2 * P2000 arrays in remote geographical locations linked over a WAN connection. I am testing this set up at present with 2 arrays in the same cabinet & same ethernet switch.

The migrating the existing workloads from P2000 is easy enough. Primary site will be running 2 ESXi hosts with HP MSA2052 connected with direct attachment through FC. The VMs on the hosts will be approx 11Tb in size total, genearting approx 10Gb of change daily. We have appox. 21Tb of usable space with 2 * 800Gb SSDs used as performance tier

The intention is to use 2 * iSCSI ports for remote replication between primary & secondary site. Been advised to use virtual replication & do not want to affect the 'on-line day' will utilise scheduled replications 3 times a day.

A number of parameters are confusing me in setting up the replication set.

Do I want to choose Discard or Queue Latest? (Queue latest seems to be the default).

Do I want a Secondary Snapshot history? If so do I need a retention snapshot history of greater than 1? Do I want to keep Primary Snapshot history? & what should be retention priority - I am guessing Never Delete would not be good option if I want to preserve disk space!

Primarily the remote replication will be used for DR site recovery & not recovery of VMs from point in time.

Second last question - I am guessing that the Enable overcommitment of pool setting should be set as ON, other internal snapshots may grow & we run out of space.

Last question - In testing scheduled replication - continually get these errors.

Scheduler: The scheduler was unable to complete the replication. - Command is taking longer to complete than anticipated. (task: du-msa-01, task type: Replicate)

EVENT ID:#A2238

EVENT CODE:362

EVENT SEVERITY:Error

Normally happens 20 secs after replication completed successfully.

Firmware version is VL270R001-01

Many Thanks

Allan Clark

[SUBHAJIT KHANBARMAN_1]

Do I want to choose Discard or Queue Latest? (Queue latest seems to be the default).

Ans: It's purely your call. When any new replication request comes and already one replication going on that time you need to decide what you want to do like you like to discard this new replication request or Queue it for later. Anyway the limit is only 1

Do I want a Secondary Snapshot history? If so do I need a retention snapshot history of greater than 1? Do I want to keep Primary Snapshot history? & what should be retention priority - I am guessing Never Delete would not be good option if I want to preserve disk space!

Ans: Again this is your call.  you can set retention count upto 16. You need to decide if you want to keep Snapshot history both primary and seconday or secondary only. retention priority never delete not good in terms of space management.

Second last question - I am guessing that the Enable overcommitment of pool setting should be set as ON, other internal snapshots may grow & we run out of space.

Ans: Yes it's always recommended to set overcommit as enabled that means volumes will be Thin volume otherwise space management will be difficult because if overcommitment is disabled then all will be fully allocated objects

Last question - In testing scheduled replication - continually get these errors.

Scheduler: The scheduler was unable to complete the replication. - Command is taking longer to complete than anticipated. (task: du-msa-01, task type: Replicate)

Ans: It depends upon size of the volume, replication schedule. We need to give enough time between consequitive replications so that data copy should happen properly.

Hope this helps!
Regards
Subhajit

I am an HPE employee

If you feel this was helpful please click the KUDOS! thumb below!

***********************************************************************************

[AllanClark]

Many Thanks for all your replies. Has put my mind at rest for this work.

On the last question. Just a bit concerend as I am currently testing with the 2 MSAs replicating on local area network with only 4 * 1Tb volumes & rate of change is not much.  When this goes into production the 2 MSAs will be replicating over a WAN & will have 4 * 4Tb LUNs.

The worklad will be 12 VMs as hosts will be vmware vSphere.

[SUBHAJIT KHANBARMAN_1 · Accepted solution]

I would suggest you to follow the below guide and refer section "Network requirements" (Page 14 onwards)

https://h20195.www2.hpe.com/v2/getpdf.aspx/4aa1-0977enw.pdf

This will help you for sure.

Also request you to mark the forum as resolved if there is no more outstanding query from your end on this issue.

This will help for everyone who are all following your forum.

Hope this helps!
Regards
Subhajit

I am an HPE employee

If you feel this was helpful please click the KUDOS! thumb below!

***********************************************************************************