Well, within 24 hours of starting to use Twitter I’m seeing what people are talking about with respect to scalability issues. The site crawls when it does serve anything but 404’s and 500’s. Twhirl’s not able to connect to the API backend. And there’s nothing on the twitter blog.
Time to do some digging to figure out wtf’s going on.
Update
It would appear that it’s slowly coming back up. But it’s not just something that’s affecting the web interface. Seems like the message routing that’s going on in the backend is not able to serve messages to the gateways (my terminology for the architecture).
Anyone out there know how I can get Twhirl to give me verbose error messages when request fails?
Update 2
Looks like someone from twitter finally responded to a developer thread about the service being down:
I can confirm that we have been down for some time due to a massive unexpected cache invalidation. We’re working to bring the site back up, although some features will be limited until caches have repopulated.
You can find the entire thread here.
Incidently, my new best friend is IsTwitterDown. Brilliant.
0 Responses to “Twitter Go Crash”