It never ceases to surprise me how some hosting companies really don't have their act together. I've had rather bad experiences with various companies in the past, so what I'm describing here isn't particularly new. Still, let me share something I experienced today as an example of how to do things wrong as a hosting company.
Our server at udedi went offline last night at around 01:00 CEST. With offline I mean completely shut off from the internet, no ping, no services, no nothing. Can happen. In these cases one would expect to be notified of the problems with details of what happened and an estimation of when we can expect the server to be online again. But nope, nothing. Even their administration interface did not indicate that there was anything going on. So around 12:30 (that's 11 hours and 30 minutes after the server went down) we opened a ticket asking about the status. 46 Minutes later we got the following reply:
Our server at udedi went offline last night at around 01:00 CEST. With offline I mean completely shut off from the internet, no ping, no services, no nothing. Can happen. In these cases one would expect to be notified of the problems with details of what happened and an estimation of when we can expect the server to be online again. But nope, nothing. Even their administration interface did not indicate that there was anything going on. So around 12:30 (that's 11 hours and 30 minutes after the server went down) we opened a ticket asking about the status. 46 Minutes later we got the following reply:
We are currently experiencing some issues but are working on it.
That was... very informative. Still no idea what's going on and no idea when the issues will be resolved. Not the kind of reply I'd expect from a professional hosting company. Worse yet, the response somewhat implies that they were aware of the problems. So why weren't we notified?
Fast forward to 19:20, after a total downtime of about 17 hours and 20 minutes, the server appeared to be reachable again. At 20:28, an hour later, the ticket was closed with the following message:
The server should be back up now. The error was basically human error where the announcement of the particular part of the network was miss-announced, and this caused servers to be up but unreachable.
This also happened at a time of day where we were unable to reach the right people quickly enough.
We got it wrong, simple as that. We will of course refund you downtime credits.
I'm glad to see that things were resolved, but... you know, it really doesn't leave a good impression. On a related note, I have no idea yet what those "downtime credits" are. The future will tell. Probably.
