New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Reefer?
On a serious note, I agree with @itrmike that it will just make the problem worse.
Welcome to why I stopped doing it and just started kicking these type...
On the other side though, if you do increase the allocated amount of ram then you will indeed see a drop in disk i/o but how much is to be seen... probably not a huge amount but if you're willing to increase customer satisfaction then by all means go for it. Ram probably isn't your highest cost and if you're like some hosts then you'll undersell and then leave room to allocate ram to customers.
Kicking people off for it is quite harsh unless they are excessively doing it, I mean there are alternative solutions such as limiting the i/o per vm. It's more about damage control to be honest, a combination of increased ram and limited i/o would result in a much more stable environment... using a san also has its drawbacks. I assume you're using it to easily manage containers and deploy them on whatever node has the lowest load? How about moving to local storage on the nodes and using the sans for backup purposes instead, if anything happens you can pull a backup onto a node and run the ct there. I don't really think sans provide many other benefits and are usually a huge waste of money for most purposes.