Digital Ocean won't let new customers create resources in four DCs, won't say why
Can't even spin up the compute necessary to reply
Digital Ocean has cloud capacity issues and won't say what the problem is.
The junior cloud company's availability matrix, which lists resources customers can deploy, was last updated on July 18 and offers the following warning about four datacenters:
Due to limited capacity in AMS2, NYC2, SFO1, and SFO2, resource creation is disabled in those regions. However, users who have existing Droplets in those datacenters can create additional Droplets there.
AMS2 is in Amsterdam's Netherlands hub, while the NYC and SFO datacenters can be found in New York City and San Francisco. The matrix indicates there are nine other datacenters – including two in NYC, two in Europe, one in the UK, and one in Singapore – with capacity for Droplets. That's the company's term for a VM. So 30 percent of Digital Ocean's datacenters have barriers to Droplet creation.
The company also offers cloud storage and other services. It is unclear if the reference to "resource" creation impacts those services, too.
The Register spotted the advisory above in early September and has since made multiple email inquiries to Digital Ocean's designated media contact, who did not respond. We asked the company's investor relations contact, too, but again received no response. We've since tried using two social media channels to reach the relevant people.
Yet here we are, writing this story without any comment from Digital Ocean.
We therefore went looking for possible reasons for the situation in Digital Ocean's two most recent earnings calls. In the first quarter of this year, the biz mentioned [PDF] management's pleasure at landing customers that start small and then display "dramatic growth in spend … paired with the adoption of additional products in our portfolio."
Maybe that means Digital Ocean reserves capacity for existing customers?
In Q2, execs spoke [PDF] of how Digital Ocean's $100 million annual hardware spend means the company has pursued, and won, good prices from manufacturers, while its footprint in the datacenters that host its infrastructure has allowed it to negotiate money-saving deals.
To us, it sounds as though Digital Ocean is particularly sensitive on cost – trying to wring the most out of its dollars, more so than you'd expect from a cloud player. Ergo, it would prefer customers use datacenters with plenty of available capacity that has already been bought and deployed, before letting them work in datacenters approaching or at capacity.
- Digital Ocean dumps Mailchimp after attack leaked customer email addresses
- Akamai buys Linode for $900m
- DigitalOcean tries to take sting out of price hike with $4 VM
- DigitalOcean sets sail for serverless seas with Functions feature
We also found an interesting nugget by using The Wayback Machine to see if the text of the availability matrix had changed over time.
That nugget is a snapshot of the page on July 15 that excludes the AMS2 datacenter from the list of facilities lacking capacity for new customers. But an April 21 snapshot does mention lack of capacity at AMS2.
Maybe that's the impact of blockchain customers bailing, as the company reported in Q2, and a subsequent uptake of capacity by other users.
Or perhaps there's some other explanation for this situation. We're curious, because we like to explore reports of cloud capacity issues so that readers can understand which clouds are best able to serve their needs.
If Digital Ocean offers more detail, The Register will update this story. But we're not holding our breath waiting for the outfit to respond to our requests. If you're a new customer that fancies working in four of the limited datacenters, do let us know how you get on – if at all. ®