dblake15
asked on
Ha to offsite location
Can you use ha to go from your main datacenter to an offsite location without srm?
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
look @
http://www.brocade.com/downloads/documents/awards/storage-magazine-feb-2011.pdf
Read "Replication revisited Once an expensive option, replication is now available in many forms
and is more affordable and effective than ever." BY JACOB GSOEDL
http://www.brocade.com/downloads/documents/awards/storage-magazine-feb-2011.pdf
Read "Replication revisited Once an expensive option, replication is now available in many forms
and is more affordable and effective than ever." BY JACOB GSOEDL
let's keep in mind that replication is just part of the solution... Even with SRM, you have to have some kind of replication in place before you can failover to the replicated instance(s).
SRM is a DR solution where HA is, as its name implies, a high availability solution. These are similar things but unique in that DR implies a second site with considerable distance between the two sites and HA is not intended to work over long distances. So, the short answer to the original question is... probably not, because you need shared storage and network connectivity (which should be a given but I'll list it here anyways) between the two sites.
If these responses haven't answered your questions, please let us know if you are looking for a DR solution or an HA solution and we'll give you more information.
SRM is a DR solution where HA is, as its name implies, a high availability solution. These are similar things but unique in that DR implies a second site with considerable distance between the two sites and HA is not intended to work over long distances. So, the short answer to the original question is... probably not, because you need shared storage and network connectivity (which should be a given but I'll list it here anyways) between the two sites.
If these responses haven't answered your questions, please let us know if you are looking for a DR solution or an HA solution and we'll give you more information.
SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Sounds like you have the bandwidth for synchronous replication of data. HA won't really work (correctly anyway) with the data on two different SANs. SRM would be my tool of choice in your environment - it will automatically failover your vms in the case of an outage on one end.
Whoops - disregard, I responded to lrpage as it were an author comment...
BTW HA does work because the two sans are synchronous so they have the same data. No need for SRM. For routine outage use maintenance mode and all vm's vmotion away, unexpected outage need to tweak the host isolation response in HA. We have had host machines go down and by the time the users realize there is an issue and start to call us the HA has done its thing and the vm's have restarted on the working host and things are back to normal
have you tested the SAN portion of this? I've seen VM's keep running even though the SAN has crashed or a datastore/LUN was taken away, so I would be leery of relying on HA to trigger for SAN issues. I thought I had heard that this is supposed to be added in vSphere 5, or something like it at least.
We have two sans one in each location, both synchronous and setup for mulitpath with esx. If one san goes down the esx hosts see this and move to the other. It is not a perfect move. takes about 5 pings and poorly written databases do not like it and they have been known blue screen. Exchange and most sql databases will complain it cannot see the data and then after a few seconds say "oh there it is and have no issues" A clean shutdown of a SAN avoids this. We have been running this for 4 yrs. This is a feature of the SAN working with ESX.
Is it perfect.. no but it has given us a very high level of redundancy and the ability to lose a building and keep running.
Is it perfect.. no but it has given us a very high level of redundancy and the ability to lose a building and keep running.
http://www.ibmsystemsmag.com/aix/storage/software/Disaster-Recovery-x-3/?ht=
Metro Mirror is generally considered a campus-level solution, where the systems are located in fairly close proximity, such as within the same city. However, the distance supported will vary based on the write intensity of the application and the network being used. In general, with adequate resources, most customers find up to a 50-kilometer distance acceptable with some customers implementing up to a 300-kilometer distance.
With Global Mirror, the target site may trail the production site by a few seconds. The frequency of creating a consistency group is a tunable parameter, and you?ll need to balance your recovery point objective with the performance impact of creating a consistency group. Many customers find a three- to five-second consistency group achievable (i.e., in a disaster, you?d lose the last three to five seconds of data).