alexei@cimedia.com wrote:
+--- In a previous state of mind, "Keith Brown" keith@netapp.com wrote: | | >or can I tag | >the failover loops as having a lower priority than the "local" loops? | | What exactly do you mean by "local loops"? If by "local loops" you mean FCAL | loops that are private to an individual filer within the cluster, then I'm | not entirely sure that we support that type of configuration anyway. In a
Ok, let me try and explain what I meant. Each filer would be primary to a disk array and secondary to another. Now, in failover a filer would become primary for both sets A & B where A is its primary. Can I tell this filer that requests for the B (secondary) array are to receive less priority than requests to A?
The reason I ask is that, in my case, the data on one array will be more important than the other. Lets say a /usr/local export vs the html content being served (or the user mail spool or financial data, etc). To me the /usr/local data is less critical than the other stuff, so I would rather the filer limit how much time/resources it spends servicing these requests.
Think of it as tagging the data with quality of service parameters.
There are no plans to add a feature like this to our Clusters, but I will take this as a request for an RFE to add such a feature.
| thing...) but I guess it does get kinda interesting when you've got three | loops to worry about. I'll have to leave this one to someone at NetApp who | knows exactly how we recommend this work physically.
When we ship more than 8 shelves with a filer we have to put the filer into a system cabinet in order to pass emissions requirements. When we have a 24 shelf cluster system in cabinets we use a 2 meter filer to shelf cable to make the jump to the third bay. I would expect to eventually see this 2 meter cable show up on our price list as an option.