The message being sent to customers claims that they “remain more aggressively priced than our peer providers”, which is nonsense as far as I can tell—unless they now consider their “peer providers” to be the likes of AWS, Azure, and GCP?
Hetzner Cloud’s price is ~40% of Linode’s new price[0], OVHcloud ~50–60%[1], and Vultr ~80%[2]. Of the big-name VPS providers, I think only DigitalOcean is more expensive now[3], and so they will be the same after this change.
What is really weird about this, they are now more expensive than comparable systems on AWS Lightsail. (Assuming they have not already updated the pricing page on linode.com)
To be fair, though, the lightsail instances that are subject to oppressive throttling aren't really a good comparison. Those should be dirt cheap, as they are useless if you use them in almost any real fashion.
A major difference about Digital Ocean (A long time user) why you may want to mindfully avoid them:
- DO intentionally does not let you easily download your disk images for backup or DR easily - which Linode and others do without issue to increase lock in. No need to reply with linux commands on how to do it. :)
- DO has a catastrophic flaw in their infrastructure where they will fail to charge a valid credit cards on file and proceed to delete all servers and backups automatically, and irrevocably.
They should not be trusted with anything in production that can't be mirrored to another cloud.
Yep - I just migrated my gaming server to Hetzner a few weeks back. With this increase in price I'll be pushed to migrate my other self-hosted services there as well.
End of an era for me - I've been with Linode for over a decade due to their pricing and positive customer support experiences. And, truth is I haven't used their support for 4-5+ years so... easy decision.
Silver lining here - Hetzner's US datacenters couldn't have opened at a better time!
Same, I have been using Linode for a little over 10 years. The main barrier to migrating off is that I will lose the IP I have been using to send mail for the last 10 years. I guess big email is going to win and I'll have to use fastmail or Google in some way now.
OVH had a data center burn to the ground just a couple of years ago, due largely to owner error. I’m not in the market for a European data center but that would give me substantial pause.
> Akamai faces the same macroeconomic headwinds of growing costs for data center space, energy, and hardware as other providers. We resisted making any changes to our offerings for as long as we could.
Can we just stop pretending that everything a buyer tells you in the first year has any relationship to reality?
Stop being surprised. You’re being lied to. You know it, I know it, everybody knows.
They always change things for the worse in the second fiscal year. And unless they bought the company at a discount they always will. The old owners cashed out. They took the money out of the company, and that money needs to be replaced from somewhere. Which means customers and employees are getting bad news.
Huh. Well damn, so they did. The acquisition didn't hit my radar till the name change.
Admittedly not a Linode customer. DigitalOcean did me a solid about 5 years ago (support gave me a credit voucher to cover a month when I couldn't afford to renew in time to keep things online) and I've had no reason to look elsewhere. Fingers crossed their recent corner cutting doesn't ruin things.
I've been a Linode customer for almost 15 years. As far as I remember, this is their first price increase ever. No surprise it happened post-acquisition :(
Same here. One of the things that made me a loyal Linode customer was that often they would have a resource upgrade (i.e. "we've doubled your RAM!") without a price increase. Not pairing this with some kind of resource upgrade feel anti-Linode. Thanks Akamai...
I have been using Linode for a little less than 15 years.
Linode's development has slowed to a crawl since this acquisition unless you start considering "Network updates to connect existing core sites to the Akamai backbone" and the likes as development.
I've been a customer since 2011 (moved away a couple of years ago). Even though they didn't increase their nominal prices, my prices raised steadily during all that time. Usually with some minor amount of extra resources added without me requesting them.
Also, most of the time, reverting the increases would take renegotiating things with support. Recently, it changed into requiring destroying my machines and creating new ones from scratch, what besides an improvement, is still as hard as migrating to another provider.
Anyway, I was not entirely dissatisfied with them, but this one is a sleazy behavior.
> Even though they didn't increase their nominal prices, my prices raised steadily during all that time. Usually with some minor amount of extra resources added without me requesting them.
This is strange, I've been a customer for a long time and never noticed that.
Prior to the acquisition they killed their long-lived “pay yearly and get a 2-month discount” offer. I loved that. With that gone and the recent increases I’m going from paying us$200/year a couple of years ago to us$288, before tax, and this with no increase in capacity.
I've been a customer of Linode since about the time they were created. I've also been a customer of Akamai in the past. Based on my experience with Akamai I was expecting this to happen just was not sure when it would.
My personal strategy to mitigate cost spikes is to have accounts with numerous VPS providers as that does not cost anything and to re-balance my nodes based on uptime, performance and cost. I found that an unexpected benefit to this method makes it easier to mitigate VPS region outages and forces me to use better automation practices even with my silly hobby sites.
This is probably not the answer you are looking for but I don't have a blog on this. For what it's worth there are existing blogs that talk about managing instances on multiple VPS providers with Ansible. I would actually defer to the proper devops people here that probably have their favorite write-ups on this topic since I basically fumbled my way through it. Some VPS providers have written Ansible plugins that integrate with their API's and some do not which just means spinning up nodes has to be done in their web interface ahead of time and setting up the SSH key trusts. For the ones that do that just means setting shell environment variables that contain the API token and calling their plugin. Linode has an Ansible plugin.
That's a really anti-hobbyist move. With CAD losing value compared to USD, it means I have been paying more in the last year already. As a tiny hobby developper who has almost no traffic, lower egress fees mean nothing to me and the higher pricing means I'll just kill my projects. I guess they don't care about losing loyal customers and are shifting towards entreprise. It's a bit sad to sell your business to people who have no care for your original set of values.
After more than a decade of never increasing prices and improving the hardware for free, charging $12/mo instead of $10/mo means all hobby projects will die and Linode hates their customers.
It should have been something that happens before acquisition or years after. What I dislike is the message that this increase sends. My bank account can absorb this increase, my heart can't accept that Linode changed.
I've been using Linode for more than a decade. I advise hobbyists not to use the cheapest VPS option and neither do I. Its noticeably less performant and just because you are a hobbyist or whatever doesn't mean you can't spend $10-20 per month.
In the United States, people regularly spend more than $10,000 per year on hobbies.
I'd like to point out that this comment is an assumption and isn't backed by any hard data.
Anecdotally, I'm a hobbyist and I've had to use higher price tiers throughout my entire time with Linode. The low-end VPS instances are insanely limited if you're looking to accomplish any sort of real work.
Yep, as a 1GB Shared Nanode customer, that was the first thing I checked. Smart move. As a hobbyist I'm pretty price sensitive, and will not hesitate to abandon them if they go too far over their $4 competitors.
Not necessarily. There tends to be a sweet spot for performance and I expect the lowest hosting plans mostly serve to acquire customers in the expectation many will upgrade to more expensive plans
Yeah, it's really sad. I am going to be switching off of Linode after using it for many years happily.
I am considering trying to exclusively use Oracle Cloud's "always free" tier. They literally give away small VMs for hobbyists and I've had one around for a few years now. Reliability isn't as good as Linode (what do you expect for free?) but as a small timer/hobbyist I don't really care.
(as an aside, driving hobbyists away from your product to Oracle of all things is quite funny to me)
I know Moore's Law isn't a thing anymore, but conventional wisdom has always been that hosting prices should go _down_ over time, due to compute, memory, and storage getting denser and cheaper, plus competition in the marketplace. This seems to not be happening anymore, except for cloud providers switching to ARM64.
Hosting providers typically make _some_ profit marking these up somewhat at deployment time, to pay for all of the surrounding infrastructure like real estate, power, racks, networking, personnel, etc. But their cash cows are customers who buy a certain plan at a certain price and then stay on those plans for years and years. Even if there are newer, cheaper plans that would work just fine, it may not be worth the time and cost of migrating. ("If it ain't broke, don't fix it.")
I once worked at a hosting provider and noted that we had a handful of truly aged VPS hosts that kept losing RAID disks and the occasional power supply. They were running a newer version of cPanel but almost everything was close to a decade old. I noted that we could replace all of these with one beefier box that would take up 1/10th the space and power but the exact words of the owner were, "Why would we replace those? They paid for themselves in the first year and are almost pure profit at this point. We won't decommission those until they die completely or customers stop paying for them."
It does get cheaper. The problem is your needs grow in lock step with the growth in CPU & memory.
The $5 linode instance is both way cheaper in real terms and way more powerful than the $5 linode instance 10 years ago.
1 modern vCPU, 1GB of modern RAM, 25GB modern SSD would be enough to host a large website 10-20 years ago. Now, its basically useless because of software bloat.
As a small business running like 30 Linode instances, this isn’t “hit the ejector seat button” stuff but it is absolutely “move up the schedule to check-in with all other cloud providers RE: pricing” stuff.
Akamai doesn't have a reputation for being interested in the bottom of the market. They announced this price increase with 31 days' notice. Future price increases and policy changes can be expected to be announced on a similar schedule.
Do you really want to wait until they announce a "hit the ejector seat button" change and you only have 30 days to make the move?
It's a publicly traded company. Of course they're going to stop all that expensive innovation and screw their customers as much as they can get away with. It's the American way.
Readit News
Hetzner Cloud’s price is ~40% of Linode’s new price[0], OVHcloud ~50–60%[1], and Vultr ~80%[2]. Of the big-name VPS providers, I think only DigitalOcean is more expensive now[3], and so they will be the same after this change.
A “bold new approach to the cloud”[4], indeed.
[0] https://www.hetzner.com/cloud
[1] https://us.ovhcloud.com/vps/compare/
[2] https://www.vultr.com/pricing/
[3] https://www.digitalocean.com/pricing/droplets
[4] https://www.linode.com/blog/linode/a-bold-new-approach-to-th...
https://aws.amazon.com/lightsail/pricing
So I am really not sure how they are looking at their "peers"
- DO intentionally does not let you easily download your disk images for backup or DR easily - which Linode and others do without issue to increase lock in. No need to reply with linux commands on how to do it. :)
- DO has a catastrophic flaw in their infrastructure where they will fail to charge a valid credit cards on file and proceed to delete all servers and backups automatically, and irrevocably.
They should not be trusted with anything in production that can't be mirrored to another cloud.
Appreciate some of the other links.
End of an era for me - I've been with Linode for over a decade due to their pricing and positive customer support experiences. And, truth is I haven't used their support for 4-5+ years so... easy decision.
Silver lining here - Hetzner's US datacenters couldn't have opened at a better time!
Deleted Comment
Bro it's only been 2 weeks
Stop being surprised. You’re being lied to. You know it, I know it, everybody knows.
They always change things for the worse in the second fiscal year. And unless they bought the company at a discount they always will. The old owners cashed out. They took the money out of the company, and that money needs to be replaced from somewhere. Which means customers and employees are getting bad news.
Admittedly not a Linode customer. DigitalOcean did me a solid about 5 years ago (support gave me a credit voucher to cover a month when I couldn't afford to renew in time to keep things online) and I've had no reason to look elsewhere. Fingers crossed their recent corner cutting doesn't ruin things.
Edit: other comments imply they’ve been improving the provided specs, so my comment might not be relevant.
Deleted Comment
Linode's development has slowed to a crawl since this acquisition unless you start considering "Network updates to connect existing core sites to the Akamai backbone" and the likes as development.
Also, most of the time, reverting the increases would take renegotiating things with support. Recently, it changed into requiring destroying my machines and creating new ones from scratch, what besides an improvement, is still as hard as migrating to another provider.
Anyway, I was not entirely dissatisfied with them, but this one is a sleazy behavior.
This is strange, I've been a customer for a long time and never noticed that.
My personal strategy to mitigate cost spikes is to have accounts with numerous VPS providers as that does not cost anything and to re-balance my nodes based on uptime, performance and cost. I found that an unexpected benefit to this method makes it easier to mitigate VPS region outages and forces me to use better automation practices even with my silly hobby sites.
We need a new independent VPS provider.
I think you're being a bit dramatic.
In the United States, people regularly spend more than $10,000 per year on hobbies.
Anecdotally, I'm a hobbyist and I've had to use higher price tiers throughout my entire time with Linode. The low-end VPS instances are insanely limited if you're looking to accomplish any sort of real work.
I am considering trying to exclusively use Oracle Cloud's "always free" tier. They literally give away small VMs for hobbyists and I've had one around for a few years now. Reliability isn't as good as Linode (what do you expect for free?) but as a small timer/hobbyist I don't really care.
(as an aside, driving hobbyists away from your product to Oracle of all things is quite funny to me)
I called it in the name-rebranding thread here.
I've been noodling around on how I'm going to move, but it looks like I should really work harder on getting that done.
Hosting providers typically make _some_ profit marking these up somewhat at deployment time, to pay for all of the surrounding infrastructure like real estate, power, racks, networking, personnel, etc. But their cash cows are customers who buy a certain plan at a certain price and then stay on those plans for years and years. Even if there are newer, cheaper plans that would work just fine, it may not be worth the time and cost of migrating. ("If it ain't broke, don't fix it.")
I once worked at a hosting provider and noted that we had a handful of truly aged VPS hosts that kept losing RAID disks and the occasional power supply. They were running a newer version of cPanel but almost everything was close to a decade old. I noted that we could replace all of these with one beefier box that would take up 1/10th the space and power but the exact words of the owner were, "Why would we replace those? They paid for themselves in the first year and are almost pure profit at this point. We won't decommission those until they die completely or customers stop paying for them."
The $5 linode instance is both way cheaper in real terms and way more powerful than the $5 linode instance 10 years ago.
1 modern vCPU, 1GB of modern RAM, 25GB modern SSD would be enough to host a large website 10-20 years ago. Now, its basically useless because of software bloat.
Do you really want to wait until they announce a "hit the ejector seat button" change and you only have 30 days to make the move?