Unsolved
This post is more than 5 years old
25 Posts
0
536
June 1st, 2015 00:00
RecoverPoint inquiries
Hello Everyone,
One of the customers is considering RecoverPoint EX (latest code) to replicate Few VNX boxes between two sites (two VNX in each site).
I'm trying to look for the documentation to find some of the few points, but I wasn't been able to find sharp answers, I hope you can help me with the below inquiries (or at least direct me to the document having these answers):
1-Can the MirrorView be used along with RecoverPoint replication ? Could you confirms if the below is correct ?
- The SP-A/B ports using RecoverPoint and MV should be different,
- You can’t replicate the SAME LUN with both technologies; you have to use either MV or RP)
Is this correct ?
2-How much capacity (Journal) should be dedicated to replicate a LUN between two sites ?
3-Can the remote replication be done without dedicating Journal LUNs?
4- Regarding the journal LUNs to be created, do they have to be created from RAID groups ? or can they be created from a pool ?
5-What's the pros and cons of having (Virtual vs physical) RPA ?
6-What happens if the link fails between the two sites ? will replication resume once link is back? or it will start from scratch ? more info would be appreciated
Your experience is highly appreciated
forshr
2 Intern
•
1.1K Posts
0
June 1st, 2015 02:00
Hi there,
1) MV and RP can be used simultaneously using different SP ports but not replicating the same LUN.
2) The replica journal has to be sized accordingly for the required protection window and underlying spindle performance to sustain peak distribution. The source journal can be a minimum size but if a failover is initiated the minimum may not be sufficient as the source becomes the replica. Check out the formula in the RP Admin Guide or use RP Calc. BCSD is an EMC internal sizing tool alternative.
3) No.
4) Either can be used.
5) The vRPA is a supported with the VNX with traditional RP or with the RP4VMs product. The difference are mainly performance which you can review in the RP Performance Guide.
6) If the WAN link fails the link is paused by RP and the RPAs start to write metadata to the source journal. Once the link is resumed he metatdata is read in order to read the data from the source array and subsequently replicate.
Regards,
Rich