Hyper-V 2012 Replica on a large cluster - many VMs to a single target CSV


hi all,

we have hyper-v 2012 cluster 5 hosts , ~200 vms (a few of them actually).

we have tried start using hyper-v replica (hvr) couple of times now, every time @ point encounter issues on destination hvr broker storage side. bandwidth required storage , start having issues merging snapshots in time. size limiting factor ~16tb size csv contain vms; csv netapp lun , cannot scale lun larger 16tb.

while potentially addressed using faster storage, more worried side effects of starting , using 200 vms single csv , lun. know there no official way split replica vms different csvs - except workaround mentioned here. workabound not suitable large clusters not managable on scale.

any thoughts, comments or suggestions welcome.


regards, leonid


hi,

from personnal experience, try keep reasonable number of vm inside csv (more cvs less vm inside). depends of storage technology use.

for large scale replica , until hyper-v allows better control on naming conventions of replica , placement , possible automate creation of replica using powershell.

here task sequence use automate replica creation.

  1. create replica , not start replication.
  2. storage migration of replica (preferably using vmm). storage migration process create folder using vm name, handy identify vm on csv.
  3. start initial replication.

in case, live storage migration allow distribute vm after replica started. have create script able decide best csv use (you can use vmm rating automate placement).

hope help.

ced.



Windows Server  >  Hyper-V



Comments

Popular posts from this blog

some help on Event 540

WMI Repository 4GB limit - Win 2003 Ent Question

Event ID 1302 (error 1307) DFS replication service encountered an error while writing to the debug log file