So, there was a power outage in Provo Utah, where I host SL/live, which took down the server again. Because the CPU is throwing errors on self check, I was unable to get it up and running again - I believe this time it was around 20-30 boot attempts before it came back up.
Right now I am on the fence about the best way to proceed - I can purchase another socket AM2/AM2+ processor to keep it running a bit longer, but most (did a few hours of research last night) are shipped from China and would take a few weeks. I have a computer coming available soon (ordered the replacement earlier this week) which is a Zotac MA-320, which would have a slower processor, but probably be adequate for hosting SL. Alternatively, I can purchase a new motherboard and transfer my current processor over to SL (socket AM3, so the current board doesn't support it) - though then I might have to purchase additional RAM, etc.
Cloud hosting at this point is also an option, something like the cheap tier of https://www.linode.com/pricing would probably be about the same as the new hardware over a couple of years, though I'm not sure just how much traffic SL generates. (Generating too much traffic would probably be a good sign, and an indicator that we could probably afford a higher tier anyway... ).
I'll post again in this topic once I've decided. If the server goes down again somehow I might have to dedicate my desktop to hosting it in the interim, though I'd rather not .
|