I am not looking at high loads nor five nines, but I am interrested in something to keep downtime short when disasters fail, like a raid controller breaking or for whatever reason physical hardware goes down, or upgrades going on, to be able to boot the vps within minutes.
In a traditional shared hosting environment, if a significant hardware failure strikes, you can usually get your accounts recovered and up on a new server within hours (if the company is good).
What to me sounds like a good way would be something along the lines of what vmware can do, a "farm" of servers in a group, using iscsi (or AoE) storage - the virtual machines can be set to automatically reboot on hardware issues for its currently assigned physical system - this way physical servers can be upraded, storage expanded/upgraded - all without downtimes longer than a reboot...
is there any vps hosters out there that provide this kind of level of service? (without a custom setup for myself, I cant afford that)
Well, I dont want the hassle of managing replication and redundancy - I think something like what these guys have would do the job: thegridlayer dot com (this forum wont let me post a url since I'm a noob here)
...maybe... is there any others with similar setups?
Funny to read that. I always thought that running a web hosting pre-suppose having sms alerts, monitoring systems and so on, but seems that not all of our competitors feel the same way. Did you had a bad experience in the past?
Anyway, forget about vmware, this software does not have the features for running a VPS hosting business, and it's very slow compared to other solutions like Xen (or even VZ).
GPLHost:>_ open source hosting worldwide (I'm founder, CEO & official Debian Developer)
Servers & our leading control panel and our Xen VPS hosting, which are already included in Debian and Ubuntu
Available in: Kuala Lumpur, Singapore, Sydney, Seattle, Atlanta, Paris, London, Barcelona, Zurich, Israel
not really a bad experience - just over the 5+ years I've done my share of hosting,and ofcourse I've experienced a couple of hw issues and they always cause some downtime.. its not mission critical, its not worth hundreds of $ extra per month per domain, but it has some value for my end users.
I am not lookin at doing vps hosting business either, I just host a a few sites for my clients and stuff, and I want a vps (or whatever) that will "never" loose my data and migrate any downed system fast (terminalgogy may not be correct here)
I dont really care what the hosters I purchase from are running, as long as it is a stable *nix system, and any "migration" and boot is quick and painless. (Very opposite of what e.g. cpanel is, its a pain with most control panel systems when you have to do some things they dont really support, moving accounts between servers usually sucks unless the sites are very plain).
From a systems and network admin perspective I have seen what vmware can do, although slow, its pretty effective at automatic and painless "failover". I just want that functionality beneath the layers of what my simple and single machine will look like to me... so far the only one I have seen with affordable (less than $500/mo) plans are TheGridLayer...