HPE Storage Users Group

Hi all,

Our infrastructure is like this :
2 ESX 6.0 U3
2 3PAR 3.2.x
4 San Switch
2 fabrics

Between each 3PAR we have made a remote copy link, Two luns are shown to the esx, one active whereas the other one is in standby. We don't have peerpersistence, we are using personna 11 and RR.

BUT, one of our ESX are using the standby path on some datastores, so, we are not able to mount these datastores (becasue in RO) on it.

WE have the same situation on multiple site with same configuration.

Could you help me to solve it ?

Thanks.

Don't export from mirror target unless full peer persistence configured (you don't need the witness part though).

I'm no VMware guru (that's next weeks training course), but I think there are 3 ways around this :

Don't export the target/DR volumes to the ESX host until you need to (manual failover).
Export but when scanning don't ever scan for VMFS (just dicover the devices and manually mount).
Install and configure SRM (VMware Site Recovery Manager) it'll handle everythng for you.

Thanks for your replies even if the news are not good !

We already looked about SRM but we are working on scada system and for now we are not sure that fault tolerance could be implemented.

It seems that we bought peer persistence license, is it hard to configure ?

PP for Manual Transparent Failover-only, you've done 90% of the job with Vmware. Just add path_management policy to the RC groups.

For Automatic Transparent Failover you need a third site to hold the Quorum Witness which boths 3PARs need independent IP access to so that a single site failure doesn't drop the IP connection for the other 3PAR.

_________________
The views and opinions expressed are my own and do not necessarily reflect those of my current or previous employers.

Thanks for your reply, could you give more information about the path_management_policy ?

_________________
The views and opinions expressed are my own and do not necessarily reflect those of my current or previous employers.