4 ms·
no ddos protection, lousy routers, single points of failure everywhere (on old servers that are about to die), running their racks very near electrical capacity
by drusenko 17y ago
no ddos protection, lousy routers, single points of failure everywhere (on old servers that are about to die), running their racks very near electrical capacity (too many servers load up at once, breaker trips), way oversubscribing bandwidth, not doing appropriate preventative maintenance, downtime because some tech slips up when editing routing configuration, downtime because some tech trips over your power chords (honestly, a top cause of downtime), downtime because another customer in a shared-colo environment trips over your power chords, etc. the list goes on and on and on.
and, worst of all, you're at their whim -- if they decide, for whatever reason, to unplug you, that's it. we prefer to eliminate anyone who has the ability to take our service down :)
- tptacek 17y agoWhat are you doing for DDOS protection now?
- drusenko 17y agoI don't want to give away the specific device we are using, but it's something that does everything in custom hardware and sits on the edge of the network. Believe it or not, we've had botnet attacks that took down $100k Cisco routers. Nasty :(
- ansonparker 17y agoif people are tripping over your "power chords" you ain't playing them right!
- drusenko 17y ago"Tripping over" is more a figure of speech. In shared hosting, it usually involves someone trying to install or remove a server in your cabinet and accidentally tweaking the power chord just enough to reset the power to a critical piece of equipment (like a storage array, for example).
- po 17y agoI think you need to re-read the part he put in quotes, not the "tripping over" part!
- crc32 17y agohttp://en.wikipedia.org/wiki/Power_chord http://en.wikipedia.org/wiki/Power_chord http://en.wikipedia.org/wiki/Power_cord http://en.wikipedia.org/wiki/Power_cord
- drusenko 17y agohaha, thanks :) didn't catch that