Every service, ranked by real customers

Moved from the old GoDaddy-owned hosts to Hetzner and netcup, with no surprises, rarely needed support, and netcup usually a bit cheaper.

Calls Contabo the worst host they have used in 15 years.

Moved hosting from Azure web apps to Linux on Hetzner with Cloudflare edge caching and cut the monthly cost dramatically.

Finds Contabo good value and flexible for both site and email hosting.

Runs KVM 2 plans and advises paying monthly, since one plan kept its old price while another jumped sharply on the next bill. Cancelling anytime is a plus for them.

Finds Contabo low quality, with roughly 30 percent CPU steal that forces buying more vCPUs than needed.

Uses RackNerd and checked their plan: about a year for a modest price with 2 cores, 3.5GB RAM, a 65GB drive and plenty of bandwidth, seen as roughly on par in price with a pricier Hostinger plan.

Moved servers from Hetzner to Scaleway after Hetzner raised prices by more than other cloud providers.

Has run several projects for months on an OVH 4 CPU, 8GB VPS with no problems and previously had good experience with Hetzner.

Bought an IPv6-only netcup VPS two months ago and enjoyed the low price with Cloudflare proxying, until a service needing IPv4 whitelisting caught them out.

Hosts with both Linode and Hetzner. Finds Hetzner cloud locations often full, so adding new servers can take months, and notes Hetzner has begun raising prices.

Running multiple VPS instances for production web apps and AI agents with zero downtime and fast customer support responses.

A newer OVH VPS-2 plan gave absolutely awful, CPU-starved performance for a game server, unlike their earlier VPS-3.

Finds Hostinger great for most use cases unless bandwidth limits matter, with only good experience in North America.

Support would not respond during outages and afterwards said they saw no problems on their end.

They currently run a VPS with IONOS and have no complaints, though they describe their use case as simple.

Moved 26 low-traffic WordPress sites onto one large Contabo VPS with CloudPanel in 2024, using about 9GB of RAM with object caching, and notes disk can drag under load.

Going through last year they found they appear to have been double billed, which they link to shifting billing dates.

Newly assigned DDoS-protected IP came cluttered with dozens of old firewall rules and support ignored tickets and refund requests for 17 days.

Had two Linux VPSes restart unexpectedly during the outage and could no longer log in with their passwords.

Recycled DDoS-protected IPs retained old customer firewall rules without being reset, and mounting issues after the acquisition forced a move to OVH.

Had three servers down for over 48 hours with no information from Hostwinds and plans to move to a new host.

Later netcup VPSs had weaker, slower virtual CPUs than the US ones, though they would still choose netcup for a VPS.

Used Hostwinds for a few years without issues, but their server hosting a newly released app's backend has been down for 48 hours with customer service suddenly unavailable.

Production e-commerce down 72+ hours after maintenance and support sent a canned reply without reading logs.

Happy customer but cannot honestly say netcup support is OK.

Has multiple sites and databases on Hostwinds that went offline and came back repeatedly for 24 hours with a broken ticketing system; credentials were reset and needed a support email to regain access.

Runs RAM-heavy quiet services on a 12GB Contabo VPS, noting shared CPU on the cheap RAM-heavy plans is real.

Business account setup took a week and tax is still not set correctly.

Locked out of the Hostwinds client area for a second day while needing to work on a client's website.

Paid for a server and received no credentials or reply to two tickets.

Received an unpaid-bill email despite paying on time through PayPal, then lost client-area access with all websites down and only AI chat bots on live chat.

Many-year customer who found service rock solid until recently, then had their VPS terminated without warning followed by this outage, and will not renew.

Customer of eight years whose VPS hosting three business sites was accidentally deleted by Hostwinds and restored after a complaint; then the server went down again with the client area inaccessible and no timeframe.

Could not connect over FTP or update their sites during the outage, though their existing sites kept working, and says they have had no problems in several years with Hostwinds before this.

They run many client sites on an IONOS VPS with no complaints. On the rare occasions they needed support it was great.

KVM 8 renewed at 777 euros after paying 276 the previous year, and support offered only 20% off.

Runs everything on OVHcloud EU VPS instances; availability can be limited, but they are much cheaper than Hetzner for similar specs.

They run hobby projects on a 1 euro per month VPS and report zero problems so far.

Servers were rebooted at peak hours on five minutes' notice and many never came back online.

Bought an OVH VPS at a very good price, but the account was banned without explanation and support was terrible, leaving two days to move a web app elsewhere.

Reported a demonstrated networking issue and support asked for root credentials instead of investigating.

Contabo oversells and has constant performance issues; mail server lagging.

Linode has been spectacular; phone support answered immediately.

Has been on Vultr with good experiences, though without any robust performance comparisons.

Found it too technical and the interface somewhat impossible to use, and OVH was much simpler.

Their Hostinger KVM 8 was CPU-limited 13 times in 10 days with steal peaking at 69% whenever pushed hard.

Worked smoothly without a single issue and offers a clean, focused set of cloud services.

Expected to leave netcup within weeks but found it superb: clean IPs, easy rDNS and no uptime problems.

Years of great experience; one node corruption was handled with off-site backups and minimal downtime.

Says OVH is fine until you need support, which can take 48 hours to give useful information, and the website UX is awful.

VPS running almost five years with no complaints or noticed downtime.

Self-hosts with Docker and Dokploy on a Hetzner VPS and describes the setup as smooth.

Moved to OVH after unreliable service elsewhere: solid uptime and an easy GUI.

Runs four Next.js apps on Hetzner with Dokploy and after a year has had very few issues, all fixed quickly, though it needs more initial setup and some Linux knowledge.

After trying many providers settled on Hetzner for production; nothing comes close even after two price rises this year.

Runs a custom Next.js e-commerce store self-hosted with Docker and Coolify on a Hetzner VPS; calls it incredibly fast and easy, and the shop sees sales up 20 to 50 percent.

Runs Next.js, Postgres and Caddy on a single cheap Hetzner VPS with Docker Compose, and pages are fast because the database sits next to the app.

Moved to Hetzner with Coolify and Cloudflare for price and full control, getting a cheap solution and plenty of disk for the database, at the cost of a learning curve.

Has run their small business site on Linode for a decade at $5 a month.

Three VPS have been generally fine despite one machine issue they had to diagnose themselves, and you get more than you pay for.

Moved from Hostinger shared hosting to Hetzner, which has been a solid choice so far.

OVH repeatedly gave blacklisted IPs while Hostinger and Hetzner were fine in production.

In production for a while and great, though prices recently rose for new orders.

Paid first and only then was asked for lots of sensitive information by email.

Hetzner is super cheap compared with Google, Amazon and even Alibaba.

A small business owner is leaving because of price increases and billing-date changes while still liking Hetzner and its products and support.

Hetzner has always been responsive, unlike a cheaper host with no support.

Wanted green infrastructure and would pay a premium, but the new pricing went from manageable to bleeding them dry overnight.

Finds OVH pretty solid, but says the newer VPS plans push pricing while performance takes a very big hit versus the older ones.

Prepaid a two-year Hostinger VPS 4 for about $200-250 with the signup discount, but renewal is around $500.

Hetzner servers have much better performance and price value than DigitalOcean but lack cloud services like k8s.

Runs a 4-core 8GB OVH VPS for a reverse proxy and some Python apps with no iowait problems.

Moved from DigitalOcean to Hetzner and database jobs took a third of the time on cheaper shared vCPUs.

On OVH VPS-1: support takes 24-48 hours and CPUs are older, but otherwise good for their SaaS app and WordPress sites.

Three or four years on an OVH VPS without an issue, though not doing anything advanced.

Could not SSH into a new Hetzner VPS; support said the account was locked for missing documentation that sign-up never requested.

They run a RackNerd VPS with the Enhance control panel to host close to 300 mailboxes, using yearly deals.

The cheapest Hetzner plan runs four websites, a bot and a proxy and is still overpowered.

Calls netcup the most boring provider because nothing ever happens, and low priced.

A 2 vCPU, 2.5 GB RAM, 40 GB SSD, 3 TB a month plan for 20 dollars a year is great value.

A long-time Hetzner customer moved to ServaRICA for 2-3x the cores, RAM and disk because current prices no longer justified their needs.

Root servers with uptime better than bigger cloud providers, though support is not instant.

Was switching from DigitalOcean but costs more than doubled, and only a locked-in price on one server keeps some business there.

Two outages in the first month that silently stopped networking for 15+ minutes.

Customer since 2022 with multiple servers and rarely any issues in Nuremberg.

Contabo had horrible drive latency of 70-120 ms, poor support, and the machine became unusable after two weeks so they sought a refund.

Enjoying RackNerd and finds the new control panel nice.

Tried Contabo and had a bad experience with unstable network and performance.

A server of four years just disappeared with no notification, reason or communication.

Contabo still has issues but upgraded its management platform to a much more modern one.

Had a server re-provisioned without warning when asking about an upgrade, with no apology or compensation.

Likes the netcup VPS, finds the CPU adequate and much better than a more expensive Hetzner VPS.

26 euro server reliably runs Nextcloud, Immich and more with no availability or speed issues.

Good experience with netcup and good deals during special holiday auctions.

Great RAM for the price but I/O bottlenecks on heavy database writes; fine for background automations.

Several VPSs over ten years without problems, and the few support contacts were excellent.

Server disappeared from the panel while the paid backup service stayed active and the domain showed another site.

Self-hosted GitLab with Hetzner object storage for registry and files works quite smoothly.

Reliable in their experience; when they tried to cancel their VPS XS, IONOS offered a 75% discount for staying another two years.

Large VPS plan handled a 1,000-person live streaming event and 24/7 streaming radio at max bandwidth without a single hiccup.

Runs 12 VPSes at Contabo and has had no high CPU load problems, with zero complaints.

Ubuntu 24.04 game servers run great with zero lag once UFW outbound ports are configured properly.

OVH provisioning via API worked beautifully, but fraud detection left their VPS stuck in provisioning for three days before cancelling.

New OVH VPS 2027 plan was a lot slower than earlier OVH VPSes and netcup, so they will not use it again.

Control panel broke so payment was not possible and tickets went unanswered.

A developer has run multi-purpose projects on a Hostinger KVM 2 VPS for over a year with little downtime, though they had to set CPU limits.

OVH's VPS 2027 plan in Strasbourg ran on 2014 Haswell CPUs and tested slow and terrible.

After two years and 20+ sites, Hostinger suspended their VPS after a malware incident with no human, no appeal and no support.

They have used an IONOS cloud VPS for 3 years and say it has never gone down or crashed. They find customer support only meh but the service itself very stable.

Monitored VPS with no complaints and only three minutes of downtime in two weeks.

Used Linode in the past and found them easy to do business with; plans a major app there.

Contabo at 6 dollars a month seems to work well.

Hetzner low-cost VPS are mostly sold out, with waits of months and scaling difficulty.

OVH monthly bill rose about 30 percent, so the poster moved demanding services to Hostinger.

Five years on Contabo: monthly fee doubled with no spec change, but support fixed an issue within hours.

OVH was fine for a year, then the AI boom basically tripled the monthly fee.

Fourteen years on Contabo, which got so bad in recent years the poster moved to netcup.

Contabo server frequently down, a six-hour outage with a status page showing no incidents, and a joke of a customer service.

Never had downtime on Hetzner and pays almost half of what they paid at Namecheap for much better hardware.

Uses Coolify on a Contabo VPS for many test projects and two production ones, calling the under 10 dollars a month a no brainer after being unhappy with Fly and Zeabur.

Contabo since 2017 for light use is fine; CPU steal is not shown but support moves you to another host free.

Server creation in the Cloud Panel failed 18 times in a row after two seconds while support remained completely unreachable from Peru.

Contabo is heavily overprovisioned.

VPS deleted two days before a written payment extension deadline and no backups were kept.

Contabo is cheapest with never an issue.

Their 20 euro server would cost 120 euros per month at the next upgrade tier after the price rise.

Their Hetzner VPS rose to about 25 euro a month and they could not rescale due to SSD size, and ARM64 machines in that datacenter were fully booked.

With 10 VPS already, found no new low-cost VPS available, nothing under 20 euros a month.

Calls Contabo historically horrendous and would rather pay 15 euros a month at Hetzner.

Has not noticed any downtime on their Hetzner VPS and guesses VMs migrate between hosts during maintenance.

Cost-optimized servers are sometimes out of stock so you must keep them running 24/7, but they are still good for web servers.

Upscaling from CPX11 to CPX21 now costs 37.49 dollars a month so they are moving everything away.

Managed Kubernetes on Scaleway turned out terrible compared to AWS, needing roughly a weekly support ticket and constant babysitting, with stability and scalability issues that lost some customers over twelve months.

Two servers at 89% average uptime over 18 months and 94% over the last three.

Servers on Hetzner with WireGuard, no issues and full control.

RS2000 G12 single-core performance swings between about 1100 and 1800, so considering a refund.

A dream for a personal self-managed project and very reliable.

VPS started segfaulting and throwing random errors suggesting CPU or RAM problems.

Used 13+ OVH Singapore machines; hardware failed repeatedly and support took hours.

Never had issues and always kindly helped on the phone.

Had a blocked-account issue with zero communication from Hetzner and is leaving.

They use IONOS VPS servers, which were great until recently when prices were raised and changed. They found customer service weak, as support did not know what Elementor was when asked on behalf of a client.

Account was deactivated about two hours after creating a server, without any explanation.

Very pleased with OVH as a VPS supplier: a mid-tier 8 GB plan used only 2-3 GB for more services than on Strato, with unlimited traffic.

They run an IONOS VPS plus the backup service for about 18 pounds a month and report never having had any issues.

Ran a Next.js app on a Hetzner VPS for a long time with a nightly recycle for memory leaks, and says it runs definitely faster than on Vercel.

Their CCX23 went from 39 to 101 dollars, a 150 percent increase they find unjustifiable.

Five days on netcup with major network issues every day and no answer to their ticket since day one.

Contabo boxes had the lowest uptime of any VPS provider they have used.

Keeping a CX22 at the old 5.39 euros a month.

Chose Contabo after DigitalOcean rejected their virtual card; hosts a client project there and has no complaints.

After a year on a budget Linux VPS, a same-spec Netcup server felt about ten times faster.

IONOS support and business practices are a bit annoying but nothing terrible.

After Hetzner price increases the price does not justify the performance of the box.

Windows VPS in Singapore for algo trading has worked fine for about a year and is the cheapest of providers tried.

Verifying a Hetzner account ended up costing 25 euros upfront.

First week mostly offline, then smooth for almost two years.

Support always replied within a few hours and was needed only twice in five years.

VPS since 2019 with early issues fixed by support, then stress-free apart from a recent price increase.

No recent outages on servers running for a long time.

Seven years with excellent performance and support and reasonable pricing.

Replaced their last Vultr VPS after ten years because the pricing is no longer competitive.

Tried a 16 GB, 8 vCPU Hetzner VPS for four months and found it very slow, then bought their own mini PC.

They used an IONOS VPS and found the cancellation process nightmarish both times they cancelled.

Says netcup is fine while you pay but turns difficult when you try to cancel, even after the contract ends.

Long-time OVH VPS user whose Haswell CPU showed its limits on light RAG work, so they are moving to netcup.

Very happy with a netcup RS 1000 G12 root server: excellent performance for the price, with no issues in two months.

Used Vultr for some years, very good with a network they love, though support content is thinner than competitors.

Game-server VPS froze constantly with 1-10 second lags; benchmarks showed throttled CPU and support sent generic replies.

Servers were so slow that a load a $20 droplet handled well struggled on Contabo.

Used IONOS before and likes the built-in firewall and the good customer support they received.

Great for a workload that needs no high uptime and cheaper than reputable hosts, but unsuitable for demanding use.

VPS went offline for a week and a ticket got no reply until they phoned the German support number.

No issues on a netcup root server with dedicated hardware over several months.

Over 7 years Contabo VPS machines have been upgraded and never gone out, though load time is not very fast.

Used Contabo for a few months and says performance explains the price because hardware is oversold.

A year on Contabo with no problems for a self-managed setup.

Switched away from Vultr because a 5 dollar 2023 instance now costs 25 dollars for the same specs.

A CPX41 went from 38.99 to 120.49.

Uses Hetzner: relatively cheaper and likes the UI.

RackNerd is cheap for decent specs and reliable, with data in Dublin on a static IP.

Uses the IONOS VPS XS for 1 EUR and is happy with the price.

Two years on Hetzner: a CX22 under 5 dollars is far cheaper than DigitalOcean and runs many containers, though only one public IPv4.

Hetzner prices are still competitive and quality is superb.

Hetzner told a business customer to wait 3 months for 4TB of extra storage, so the company moved to AWS and Google.

Used Contabo before and would not recommend it: performance is really bad and unstable.

A Hetzner server paid at 15.99 euros would now cost 45 euros or more.

Using Hetzner for a few months: great value and zero downtime.

Consistently great Hetzner support, with a human tech fixing a lockout mistake and lenient billing.

Long-time Hetzner customer: several reminder emails over weeks before a block, which prompts payment.

Moved their whole estate for pricing four months ago and was shocked by the new-instance increase, though owned instances stay price locked.

In their experience all Hetzner Cloud servers can reach nearly 10 Gbit/s depending on neighbouring guests, though no speed is guaranteed.

About 15 minutes of downtime across a year on Hetzner.

Moved to Hetzner a few months back and lowered costs significantly.

Multiple Hetzner cloud servers were flapping up and down every five minutes during the FSN1 incident.

Hetzner network outage with nothing on the status page; Reddit confirmed the issue first.

Affordable storage with a clunky UI but top-notch functionality.

Has never had an issue on a Contabo VPS with lots of storage at 11 a month, and support is quick.

After declining Hetzner ID requirements they moved to IONOS and so far it is good at half the cost.

Eight years on a server with no problems.

They host multiple sites on a Linode VPS, noting the cheapest plan is 5 dollars a month.

IONOS at 2 dollars a month is great for a smaller project.

A Hostinger KVM2 showed slowness and outages with over 90% CPU steal, with only a bot answering support, so they moved to netcup.

An 8 vCPU 16 GB server rose from 25 to 32 to 87 euros within a few months.

GPU instances (L40S and L4) have been out of stock very frequently for the past month, requiring hours of retries before a server can start.

Fifteen years hosting websites with IONOS/Strato and two years on a VPS without an issue.

Hostinger banned their VPS for abuse, but the compliance team was knowledgeable and restored the account once the trigger was found.

Has had two servers for 6-7 years and the price hike might make them cancel.

Cancelled a CPX62 at 60 dollars to build a home server; it would now cost 160.

Moved from OVH and was happy with better speeds at the same price but is going back to OVH after the price changes.

Hetzner is the only provider where they have had hardware failures, plus recent volume downtime from switch firmware upgrades.

Years on AWS and Hetzner with comparable reliability and only a few hardware failures, apart from Hetzner Object Storage.

Existing CCX33 client notified that the price would go from 55.49 to 76.99 dollars on April 1st.

Existing CCX13 rose from 14.86 to 19.62 and new orders are far higher.

Warns existing VPS owners not to scale or change servers because the new prices are rough and may reach existing customers.

Did not expect a 150 percent increase for their little server.

Over a decade as a customer and first time really disappointed, calling the doubling or tripling of prices out of order.

Chose Hetzner Cloud for performance over DigitalOcean, Vultr, Linode and IONOS and has appreciated it more over 4-5 years with no payment problems.

Moved from Linode for lower prices, but at the new prices it would cost about the same, so they will look elsewhere.

New VPS prices were 2-3x higher than the week before.

Cloud-only customer shocked by a CCX23 jump from 31.49 to 85.99 euros overnight and considering bigger providers.

Found Linode and DigitalOcean to have worse disk and memory performance than equally specced Hetzner VPSes.

A CCX13 quoted at 15 dollars a month in March now costs 50.99, and their own shared VPS would rise from 6.99 to 22.99.

Recent customer expected a 10-15 percent rise, not 150 percent, and calls the unannounced hike a poor sign of reliability.

Fifteen cloud servers costing 700 euros a month would rise toward 2000 euros, so the team is moving away after almost 10 happy years.

Keeps Hetzner for rare testing and good support for API provisioning but prefers netcup for value.

CPX11 cloud servers went from 5.99 to 17.49 euros, forcing a search for another provider.

Server 1200 km away still gets 30 ms ping and they doubt cheaper providers match the quality of service.

Chose Hetzner as the cheapest solution and found reliability and service far better than expected.

Terrible customer support experience prompted a recommendation to give up and move workloads to Vultr.

Pays under 10 dollars a month for an 8 GB Hetzner VM even after price hikes, and Hetzner emailed them about a misconfigured public Redis.

Prefers a Hetzner VPS with self-hosting for Next.js projects and has found no downsides yet.

On a Hostinger KVM 8 the CPU throttle starves the VM so SSH and Docker stop answering, making recovery hard.

Nightly 50GB backups to Hetzner Object Storage fail about two nights a month; fast when it works.

Hetzner S3 in production crashed weekly with the worst performance experienced.

Hetzner S3 gives regular 504 errors even with a 250GB repo; IONOS S3 was blazing fast.

Moved from Hetzner S3 to Wasabi; Hetzner is brilliant for dedicated machines but not S3.

Reliable service since 2022 overall, though a multi-week outage on Las Vegas server backups and snapshots went unresolved by support.

Chose OVH for high DDoS bandwidth, unlimited transfer and stable performance with no CPU steal.

Hetzner S3 restore for Hyperbackup took weeks due to frequent restore failures.

Backup job failed because Hetzner Object Store was unavailable.

Switched almost all infra from Azure to Hetzner with lower prices and better I/O.

Hetzner S3 used for PBS off-site storage fails jobs about once a week.

Six months on Hetzner S3 was really bad, so they moved to Backblaze.

Loves Hetzner but its S3 product has random outages and HTTP timeouts.

Hetzner S3 for Terraform state and static hosting behind a cache has no issues.

Images served from Hetzner buckets regularly failed to load, leading to angry users.

Followed a Hetzner VPS setup tutorial and says everything works perfectly.

They have used Contabo for 4 years across 5 different VPS and found it the cheapest option.

Availability is spotty so they cannot add another node during a spike, and are considering dedicated servers.

They moved their sites from IONOS shared hosting to an IONOS VPS and are really happy with it, with the trade-off that they now handle much of the maintenance themselves.

Has used the Hetzner plus Cloudflare setup for almost every site they have built and finds it easy and cheap.

GitHub Actions runners at about 10 percent of the cost with better performance.

Happy long-time Hetzner customer running the cheapest VPS as a WireGuard proxy with no problem.

They recently got an OVH VPS in the Canadian datacentre and say service has been great so far, though long-term reliability is unknown.

Moved Next.js production apps to Hetzner two years ago after platform outages elsewhere and reports solid uptime and performance for the price.

They moved one project to Hetzner with no issues and got an 8 core server for $20 a month within minutes, noting a two-machine limit for new users.

They switched from Hetzner to OVHcloud for HIPAA compliant data centers and got a more powerful VPS for less money, though the admin interface is less nice.

A 1 vCPU, 1 GB RAM, 10 GB plan hit 80 to 100 percent CPU under WireGuard relay traffic of about 230 Mbit/s, which another user saw at only 15 percent; throttling is suspected.

They run RackNerd's 5 core, 6GB RAM, 100GB SSD VPS for $40 per year, bought on Black Friday. They say they have to deal with a lot of CPU steal time.

They say IONOS has been good for them in this VPS thread.

They use a netcup VPS close to their location for real-time applications needing low latency. They say it works perfectly fine with no issues.

Root Server with dedicated cores runs at nearly 100% CPU almost constantly, pausing only briefly between tasks; as far as they know, 24/7 full load is allowed.

Support was helpful and the experience pretty good overall despite the reputation.

Moved from Vercel to a Hetzner CX33 with Coolify, found setup straightforward, much faster than Vercel and about 7 dollars a month.

Paid for Sydney but got Singapore routing, wildly variable speeds and support insisting otherwise.

Contabo VPS dropped off the network every two hours; another node was taking the IP and support asked for the root password.

Calls OVH incredible: zero CPU steal on the VPS, great support and ridiculously fast NVMe disk for the price.

Moved from Contabo to Hetzner three years ago with zero downtime or network issues since.

Contabo support is the worst experienced.

Contabo VPS renewal price is almost double the new-buy price.

Contabo VPS 3 SSD price was raised to over 30 dollars a month, so the poster is leaving.

They had a good deal on a legacy VPS package that IONOS honored for a while, until it stopped. They also had trouble stopping the charges and moved to Namecheap hosting.

With 25 cloud VMs they had only two outages in a year, a switching problem and a roughly one-hour volume issue.

Eight-year customer sees storage, load balancer and unreachable cloud machine issues, but support replied in 15 minutes and fixed it in 25 despite no SLA.

Verified ID and card and two days later got an email that the account was closed.

They pay 6 euros for a VC1S and the automatic migration would put them on a 30 euro DEV1-L. They feel Scaleway just wants to be rid of such customers.

They run some machines and DNS with Scaleway, say everything is fine and support is good.

OVH took a week to provision a small Montreal VPS, then provisioned it anyway after they cancelled; Hetzner delivered in minutes.

Serves 250k requests a month including a CMS from a 5 euro Hetzner box with under 50ms TTFB.

They have used Scaleway Kubernetes, managed Postgres and compute and been happy, calling it a good middle option between hyperscalers and Hetzner on price and services.

They have run similar WordPress booking workloads on Hetzner CX21/CX31 and found the NVMe performance made a noticeable difference, with good value versus DigitalOcean.

Finds the console inaccessible beyond a basic overview, with even firewall changes annoying.

Had to recreate CX23 servers repeatedly to get AMD rather than slower Intel CPUs.

Hetzner would not verify the account even after document verification.

Got a heavily discounted Hostinger 2c/8GB VPS at about $120 a year, but renewal is around $200.

Hetzner is pretty good and cheap.

Uses both for business: OVH cloud/VPS has much lower CPU performance than Hetzner but is cheaper.

Over ten years with OVH, from VPS up to private cloud, with kit that performs well at a fair price.

Gitea for a SaaS company runs on Contabo Singapore VPSs, cost effective with no network issues.

A RackNerd special came with a US location with unusable ping; support moved the server to Amsterdam for 6 dollars a year.

Contabo has great prices and reliable servers.

Using RackNerd, which has been great, but needed a European location.

Has two RackNerd VPS: great company, issues resolved quickly and no downtime.

Runs a portfolio and small projects on one $7 Linode instance and it has been fine.

Some servers ran 7+ years fine while others had 24-hour stops several times a year and one lost almost a day.

A week with a Contabo VPS: unstable, hard to recover and failing VNC access, then a pro-rata refund.

A VPS ran four years without trouble but running their business on Contabo's infrastructure was chaos.

Performance is not amazing but the VPS work well and fit a price point.

Upgraded from 2 vCPU/8GB to 4 vCPU/24GB and the Kubernetes migration finished seamlessly in about a day.

Could not get SSH or VNC access working during the first two weeks of using Contabo.

Seven to eight years on Contabo VPS without a failure or downtime.

Half a year in production with a single five-minute outage, roughly 99.9% uptime.

Uses the storage VPS range and says it has always been reliable with no downtime.

Chose a netcup RS1000 root server for a Minecraft server and says it is doing wonders so far.

Loves OVH for value: 8GB RAM, 75GB SSD, unlimited bandwidth and 4 cores on the base plan.

IONOS X VPS at about 2 dollars a month has pretty good specs.

One year of OVH VPS and dedicated with zero problems.

Two RackNerd servers for four months with no downtime.

Low-end performance needs very well met by RackNerd.

They run a fleet on the retired VC1 plan and object to the forced migration and reinstall, finding the replacement resources less stable for the price. They plan to move to another provider.

A 2GB OVH VPS ordered before January 2025 shows 482 days of uptime.

After the first order was refused by a security check, their second order was accepted. They call it the cheapest VPS by a long way with no complaints so far.

Hetzner support is quite good, always quick to respond and resolve; object storage and Storage Box are weaker.

Got an OVH email that their $6.00 VPS plan would rise to $8.50, a 41.6% increase.

Runs an API, MCP servers, analytics, uptime monitoring and Forgejo on one netcup VPS for 12.60 euros a month.