I am sure many people may have noticed that Tweaked for your Pleasure went offline some time in the morning only to come back online at a later time. Well, I figured I may as well write why so. Basically what happened is my vps provider called VPS.NET on Feb 8, 2012 decided for a reason I do not know to do a SAN synchronization at the database that the vps is located.
Well this was all good but what happened is the I/O on the SAN unit went nuts and well it crashed a lot of things and forced VPS.NET to ultimately having to stop the SAN synchronization and delay it until a later time.
So at around 6:30pm everything started to come back online. However that was not the end of it as all my databases decided to crash and messed up. I fixed them at around 9pm and everything is still working up to now.
This said there seems there will be more downtime coming but this is expected downtime in which I will get another system to take the load for a while. The email below is about the up and coming downtime that was sent from vps.net.
As you aware the LON-I SAN you are currently on is experiencing
slowness due to a re-sync caused by the Starwinds SAN solution. We
expect this to be complete by the end of Sunday 12th Feb 2012. However
due to this issue we are elect to bring forward our maintenance window
for replacing this SAN. As such starting next Tuesday,14th Feb 2012,
we will be migrating people from Starwinds, to a new enterprise class
HP Storage solution which offers much higher levels of performance and
stability. There will be no data loss or IP change in this process.
As part of this Windows VPS on this cloud will see a reboot, Linux VPS
will not see any downtime at all as the migration will be seamless.
Going forward this will make the cloud much more stable. Should you
have any questions or comments regarding this do please let us know by
So there you have it – that is why it went down. Oh and CloudFlare – why did you not make any cache pages?