It started at around 1:40pm AEST (Sydney time). Both of my VPS with VPSLink went offline. One of them, an OpenVZ Link-3, came back 90 minutes later. The other one, a Xen Link-4 hosting this very website, did not come back online until around 6:30pm in the evening. Pretty bad day for HostingFu and around 15 other websites hosted on this VPS.

Bandwidth graph — blank bit are the down-time.

Again, the first thing I did when the VPS went down is to search on Twitter and saw it wasn’t just me who have lost contact with his servers. VPSLink has done a good job to use this offsite media to communicate & report updates regarding the outage. Turns out to be a power failure.

There was actually some complication to bring my 2nd VPS back online. It did come back later that afternoon, but it had some problem and sshd wasn’t running. It turns out that my /dev is completely borked, as udevd wasn’t running. It did not run as I upgraded to Debian Squeeze, and its udev package does not work with para-virtualized kernel which was in very outdated 2.6.18. So I had to use Xen console to get onto my VPS, downgrade the udevd, so I can restart ssh and other affected service. However in order to use Xen console, I had to wait until VPSLink’s console panel to come back up, which wasn’t available until around 6pm…

Hopefully everything is resolved now and we will see a formal explanation from VPSLink soon.

On another note, I also managed to sign up LayerBoom‘s GeoVPS — a Linux KVM based virtual server company with its own customised panel. Very interesting product, and hopefully you’ll see a review here soon.

Update: VPSLink has posted the response to downtime on their forums.