- Joined
- Oct 2, 1998
- Messages
- 15,223
Here's the deal folks.
As you are aware, we moved from our dedicated servers to a new VMWare style setup late last year. This move was supposed to provide more reliable hosting, better support, more failover and redundancy, and all sorts of wonderful technical babble that doesn't mean anything to you guys. I was promised repeatedly that this would be a great thing, despite my reluctance.
Since the move, we've had outages daily, if not several times per day.
Initially, I was told this was because of the size of the database; because of our huge number of posts, the posts db was locking regularly, the search index was throwing problems, etc etc etc etc etc. The support guys tweaked, and fiddled, and fixed, and the search works... but the problems didn't stop.
It now turns out that this was a hardware issue all along. Apparently, despite the audit that was performed prior to our picking this hosting solution, our purchasing equipment based on the recommendations of this hosting company, and despite the reassurances that everything would be great with lots and lots of excess processing availability and such... the disk throughput is our bottleneck. At least, that's what I've been told this time. For some reason, the I/O goes into the toilet and the hosting company can't figure out why.
Needless to say, this is inexcusable. This problem should have been resolved 4 months ago. I'm not happy that you guys are dealing with problems, and I'm pissed off and embarrassed by this entire thing.
Supposedly (again) there is a plan: we are being moved to our own dedicated box instead of the virtualized machine. If this fails, my plan be is to yank the site, and move us back to dedicated servers again. If that doesn't work, who knows.
I apologize for the difficulties you guys are experiencing. I'm beyond frustrated. I've articulated how pissed off I am at a few people today, and hopefully those responsible get their heads out of their fourth point of contact.
If not, go to plan b.
Thanks for your continued support.
As you are aware, we moved from our dedicated servers to a new VMWare style setup late last year. This move was supposed to provide more reliable hosting, better support, more failover and redundancy, and all sorts of wonderful technical babble that doesn't mean anything to you guys. I was promised repeatedly that this would be a great thing, despite my reluctance.
Since the move, we've had outages daily, if not several times per day.
Initially, I was told this was because of the size of the database; because of our huge number of posts, the posts db was locking regularly, the search index was throwing problems, etc etc etc etc etc. The support guys tweaked, and fiddled, and fixed, and the search works... but the problems didn't stop.
It now turns out that this was a hardware issue all along. Apparently, despite the audit that was performed prior to our picking this hosting solution, our purchasing equipment based on the recommendations of this hosting company, and despite the reassurances that everything would be great with lots and lots of excess processing availability and such... the disk throughput is our bottleneck. At least, that's what I've been told this time. For some reason, the I/O goes into the toilet and the hosting company can't figure out why.
Needless to say, this is inexcusable. This problem should have been resolved 4 months ago. I'm not happy that you guys are dealing with problems, and I'm pissed off and embarrassed by this entire thing.
Supposedly (again) there is a plan: we are being moved to our own dedicated box instead of the virtualized machine. If this fails, my plan be is to yank the site, and move us back to dedicated servers again. If that doesn't work, who knows.
I apologize for the difficulties you guys are experiencing. I'm beyond frustrated. I've articulated how pissed off I am at a few people today, and hopefully those responsible get their heads out of their fourth point of contact.
If not, go to plan b.
Thanks for your continued support.