[dangrossman]

Jason, what's up with this comment?

Where did you find this 'complex edge case of a system' in his post? This 'unique setup'?

Why do you think 'impossible things were about to become' when he was able to easily move the sites to WP Engine... after paying a 3rd party to do it?

He didn't describe a single special requirement that I can see. Just plain old WordPress with some existing data to move. Not even a large network or a large amount of traffic. This is supposed to be WP Engine's bread-and-butter customer.

All of WP Engine's plans other than "Personal" say "Hey WordPress MultiSite customer, come here! This plan is for you! Hundreds of thousands of visits a month and 20GB+ of storage!" http://wpengine.com/pricing/

Why are you calling him a 'toxic customer'? He's been running this WordPress install himself on an unmanaged VPS. He wasn't wasting their support staff's time on unnecessary questions -- they were wasting his! That's not a 'toxic customer', that's a 'toxic business'.

I'm having a hard time connecting this comment to the HN submission at all.

[kordless]

He also didn't describe much of any of his technical requirements. I'd say it's a fair assumption his setup was enough different to cause an issue. Whether or not that's a fireable offense is up to the company.

[techtalsky]

He did, however describe the eventual problem: his comments table was larger than what Wordpress's out-of-the-box feature (Recent Comments) was able to sort. He had very overpowered hardware in his custom situation. It strikes me that this was an extroardinary but still within band problem that WPEngine should have eaten the cost for solving OR tell him MUCH more swiftly they couldn't.

[jacques_chester]

The eventual problem was just the straw that broke the camel's back.

The weeks of shitty service beforehand are what really did it.

[jacques_chester]

It's a bog ordinary Wordpress multisite installation. The most exotic thing about it was WP-supercache and WPEngine have clever inbuilt logic to safely disable that when you migrate across.

[trevelyan]

I'm more curious what VPS package you had powering your stuff at Linode and how much you were spending on it each month? Reason being that 250 USD buys a LOT of server these days and it is surprising you ended up with underpowered hardware if your VPS wasn't giving you any problems.

If you're looking to upgrade in the future, you might be much better off throwing a dedicated server at the problem and setting a nightly backup to S3. Also, thank you for posting.

[jacques_chester]

I use 2 512Mb Linodes. One for nginx/PHP, one for MySQL. Putting these on different servers is by far the biggest leap in performance I ever got.

Monthly cost is about $40 ... unless you count my time.

[wyck]

Do you not read the part about him having over 400k comments? Don't you think that is a bit outside the regular scope?

[dangrossman]

I don't think that was in the post originally. Either way, it's not outside scope, it's not a big deal, and it's not an excuse for weeks of failing to provide support. They didn't fail to even move the site because of the number of comments. Matt Mullenweg, the founder of WordPress, commented himself that they host many blogs with far more comments than that without issue.