fwiw, Bunny are the people that announced S3 compatibility for their object storage in Q2 2022 [1]
> We can’t wait to have this available as a preview later in Q2 and truly make global storage a breeze, so keep an eye out!
then apologised for missing that in September 2023 [2]
> We initially announced that we were working on S3 support for Bunny Storage all the way back in 2022. Today, as 2023 is slowly coming to an end, many of our customers continue to follow our blog, hoping for good news about the release.
changing the roadmap to early 2024 [2]
> But we are working aggressively toward shipping S3 compatibility in early 2024.
That same post also has the beautiful "At bunny.net, we value transparency." quote.
It's early 2026, and they're literally ignoring my support requests asking about what the roadmap is looking like for this now.
So, do not trust their product or leadership at all.
Yeah I'm in the same boat. I was pretty excited to bring stuff over from Cloudflare but the missing S3 compat. and the communication around that was (and still is) a dealbreaker for me.
I've been struggling with Bunny the last couple of days.
Their log delivery api is delayed by over 3 days, despite them promising only "up to 5 minutes delay" in their docs: https://docs.bunny.net/cdn/logging
Why isn't it on the status page you might ask? Oh, that's because a delay is not "critical", but I fear I am losing loglines now, their retention is 3 days.
It's an interesting strategy for them, because it doesn't inspire confidence in me about their other offerings. When they can't reliably operate a log delivery API or be transparent about issues, it's hard to trust them with something as critical as a database.
I would have no concerns around reliability uptime running my own database.
I would have concerns around backups (ensuring that your backups are actually working, secure, and reliable seems like potentially time intensive ongoing work).
I also don't think I fully understand what is required in terms of security. Do I now have to keep track of CVEs, and work out what actions I need to in response to each one? You talk about firewall rules. I don't know what is required here either.
I'm sure it's not too hard to hire someone who does know how to do these things, but probably not for anything close to the $50/month or whatever it costs to run a hosted database.
It's not about it being hard, it's about delegating. Many companies are a bit less sensitive to pricing and would rather pay monthly for someone else to keep their database up, rather than spending engineering hours on setting up a database, tuning it, updating it, checking its backups, monitoring it and making it scale if needed.
Sure, any regular SME can just install Postgres or MySQL without even setting much up except with `mysql_secure_install`, a user with a password and an 'app' database. But you may end up with 10-20 database installs you need to back up, patch and so on every once in a while. And companies value that.
The vast majority of products with paying customers need better availability than “database went down on Friday and I was AFK until Monday, sorry for the 3 day downtime everyone”
While in public preview, Bunny Database is free.
When idle, Bunny Database only incurs storage costs. One primary region is charged continuously, while read replicas only add storage costs when serving traffic (metered by the hour).
Reads - $0.30 per billion rows
Writes - $0.30 per million rows
Storage - $0.10 per GB per active region (monthly)
The best thing about their pricing is that you can prepay. So if you have a runaway cost, it can stop before you run up a 5 or 6 figure bill, unlike Azure/AWS/GCP/CF.
Adding my voice to the chorus here: they've established a pattern of introducing new features and never really getting them past the 80% point. No qualms with the CDN; it's a sweet spot among providers. But their other offerings have been frustrating me for years now.
Will give this a spin. They’re one of the few cloud-y providers that has both prepayment and a rate limiter that doesn’t charge for rate limit exceeds (still blows my mind that providers charge for blocks).
It does feel like they're spreading their resources pretty thin though, the S3-compatible interface for their file storage has been "coming soon" since 2022.
Huh, how? Did you have to modify your site a lot to do switch?
I tried to test it out as a CDN replacement for Cloudflare but the workflow was a lot different. Instead of just using DNS to put it in front of another website and proxy the requests (the "orange cloud" button), I had to upload all the assets to Bunny and then rewrite the URLs in my app. Was kind of a pain
When I tried it last year, their edge compute infra was just not there yet. It could not do any meaningful server-side rendering because of code size, compute and JS standard constraints.
I have been using them for over a year. THey have the same flow as Cloudflare, point domain to thier CDN, set CDN Pull Zone to target your server. I havent had to do anything.
They even support websockets.
Why they cant do is the TUnnel stuff, or at least fake it. I have ipv6 servers, and I can't have the IPv4 Bunny traffic go to the ipv6 only sources.
> Not every project needs Postgres, and that’s okay. Sometimes you just want a simple, reliable database that you can spin up quickly and build on, without worrying it’ll hit your wallet like an EC2.
Isn't the operational burden of SQLite the main selling point over Postgres (not one I subscribe to, but that's neither here nor there)? If it's managed, why do I care if it's SQLite or Postgres? If anything, I would expect Postgres to be the friendlier option, since you won't have to worry about eventually discovering that you actually need some feature even if you don't need it at the start of your project. Maybe there are projects that implement SQLite on top of Postgres so you can gradually migrate away from SQLite if you need Postgres features eventually?
Marek here from bunny.net. We’re not saying SQLite is universally better than Postgres. The trade-off we’re optimizing for is cost model and operational simplicity.
Even as a managed service, Postgres DBaaS still tends to push users into capacity planning, instance tiers, and paying for idle headroom. Using a SQLite-compatible engine lets us offer a truly usage-based model with affordable read replication and minimal idle costs.
Per million rows written: Bunny $0.30, Cloudflare $1.00 (first 50M/month free)
Per GB stored: Bunny $0.10/region, Cloudflare $0.75 (5GB free)
Bunny also has a lot better region selection, 41 available vs. Cloudflare's 6 (see https://developers.cloudflare.com/d1/configuration/data-loca...). Even though Bunny charges storage per region used where Cloudflare doesn't, Bunny still comes out cheaper with 7 regions selected. Bunny lets you choose how many and which regions to replicate across; Cloudflare's region replication is an on/off toggle that is in beta and requires you to use "the new Sessions API" (I don't know what this entails).
The main reason I haven't tried out D1 is that it locks you into using Workers to access the database. Bunny says they have an HTTP API.
I plan to stick with VPSes for compute and storage, but I do like seeing someone (other than Amazon) challenge Cloudflare on their huge array of fun toys for devs to play with.
Some ISPs have bad peering with Cloudflare (e.g. Deutsche Telekom). Not Cloudflares fault but it makes it a bad choice if your customers are in Germany.
Not a technical reason, but given Cloudflare's recent business practices where they hold you hostage if you don't upgrade to an enterprise plan are a pretty good reason to avoid imo.
Some European companies migrate their dependencies from US clouds to European ones. Turso is registered in Delaware. Bunny HQ is in Slovenia. Different data related policies apply.
fwiw, Bunny are the people that announced S3 compatibility for their object storage in Q2 2022 [1] > We can’t wait to have this available as a preview later in Q2 and truly make global storage a breeze, so keep an eye out!
then apologised for missing that in September 2023 [2]
> We initially announced that we were working on S3 support for Bunny Storage all the way back in 2022. Today, as 2023 is slowly coming to an end, many of our customers continue to follow our blog, hoping for good news about the release.
changing the roadmap to early 2024 [2]
> But we are working aggressively toward shipping S3 compatibility in early 2024.
That same post also has the beautiful "At bunny.net, we value transparency." quote. It's early 2026, and they're literally ignoring my support requests asking about what the roadmap is looking like for this now.
So, do not trust their product or leadership at all.
[1] https://bunny.net/blog/introducing-edge-storage-sftp-support... [2] https://bunny.net/blog/whats-happening-with-s3-compatibility...
Yeah I'm in the same boat. I was pretty excited to bring stuff over from Cloudflare but the missing S3 compat. and the communication around that was (and still is) a dealbreaker for me.
I've been struggling with Bunny the last couple of days.
Their log delivery api is delayed by over 3 days, despite them promising only "up to 5 minutes delay" in their docs: https://docs.bunny.net/cdn/logging
Why isn't it on the status page you might ask? Oh, that's because a delay is not "critical", but I fear I am losing loglines now, their retention is 3 days.
It's an interesting strategy for them, because it doesn't inspire confidence in me about their other offerings. When they can't reliably operate a log delivery API or be transparent about issues, it's hard to trust them with something as critical as a database.
Maybe I'm not the target market for this, but how hard is it REALLY to manage a RDBMS?
Any Linux distro can have MySQL or Postgres installed in less than five minutes and works out of the box
Even a single core VPS can handle lots of queries per second (assuming the tables are indexed properly and the queries aren't trash)
There are mature open source backup solutions which don't require DB downtime (also available in most package managers)
It's trivial to tune a DB using .conf files (there are even scripts that autotune for you!!!)
Your VPS provider will allow you to configure encryption at rest, firewall rules, and whole disk snapshots as well
And neither MySQL or Postgres ever seem to go down, they're super reliable and stable
Plus you have very stable costs each month
I would have no concerns around reliability uptime running my own database.
I would have concerns around backups (ensuring that your backups are actually working, secure, and reliable seems like potentially time intensive ongoing work).
I also don't think I fully understand what is required in terms of security. Do I now have to keep track of CVEs, and work out what actions I need to in response to each one? You talk about firewall rules. I don't know what is required here either.
I'm sure it's not too hard to hire someone who does know how to do these things, but probably not for anything close to the $50/month or whatever it costs to run a hosted database.
As for the CVEs: you just need to install from your OS’s package manager and run periodic updates. The communities take care of this very well.
It's not about it being hard, it's about delegating. Many companies are a bit less sensitive to pricing and would rather pay monthly for someone else to keep their database up, rather than spending engineering hours on setting up a database, tuning it, updating it, checking its backups, monitoring it and making it scale if needed.
Sure, any regular SME can just install Postgres or MySQL without even setting much up except with `mysql_secure_install`, a user with a password and an 'app' database. But you may end up with 10-20 database installs you need to back up, patch and so on every once in a while. And companies value that.
One instance maybe but multi regional?
Look into the capabilities of what I consider the leading edge of open source RDBMS managed solutions, Yugabyte: https://www.yugabyte.com
And tell me how easily you can achieve this "out of the box"
If you don't care about business continuity or high availability then everything gets easier
> And neither MySQL or Postgres ever seem to go down, they're super reliable and stable
The box they're on goes down
> The box they're on goes down
So? Not everyone needs 99.999999% availability.
The vast majority of products with paying customers need better availability than “database went down on Friday and I was AFK until Monday, sorry for the 3 day downtime everyone”
If you're offering a hosted service, I've got bad news for you.
Serverless, managed databases and even multicloud won't save you. You'll still have to be on call.
Don't want to be on call? Design your stuff so it works local first.
> If you don't care about business continuity or high availability then everything gets easier
And some do, so what's your point?
Read the entire thread to find out.
This is not a database of bunnies
That's the only reason that I clicked.
Especially after httpstatus.cat and the animal game the other day.
I've been had :(
Seconded
Accurate, but missing required punctuation of at least 1 grawlix of length 4 and 2 interrobangs
Pricing Details:
The best thing about their pricing is that you can prepay. So if you have a runaway cost, it can stop before you run up a 5 or 6 figure bill, unlike Azure/AWS/GCP/CF.
Their CDN has a minimum $1 charge.
I was testing IPv6 origin support (they don’t support it), and they billed me $2 for a couple of test requests. I was testing at the end of the month.
With other providers, this would have cost only a few cents.
That sounds concerning but I wouldn't use other providers anyway which can decide to charge me 5-6 figures due to user error or hacked account.
Adding my voice to the chorus here: they've established a pattern of introducing new features and never really getting them past the 80% point. No qualms with the CDN; it's a sweet spot among providers. But their other offerings have been frustrating me for years now.
Will give this a spin. They’re one of the few cloud-y providers that has both prepayment and a rate limiter that doesn’t charge for rate limit exceeds (still blows my mind that providers charge for blocks).
Pretty cool. I’ve been using Bunny as a Cloudflare replacement for a couple of years and my experience has been flawless.
It does feel like they're spreading their resources pretty thin though, the S3-compatible interface for their file storage has been "coming soon" since 2022.
That’s the problem with keeping a public roadmap - in the end everyone is unhappy and also not really more informed
Yeah that's true. The lack of S3 compat hasn't been an issue for me personally but it would be nice to have it for their edge storage.
I, too, have the same worry.
Huh, how? Did you have to modify your site a lot to do switch?
I tried to test it out as a CDN replacement for Cloudflare but the workflow was a lot different. Instead of just using DNS to put it in front of another website and proxy the requests (the "orange cloud" button), I had to upload all the assets to Bunny and then rewrite the URLs in my app. Was kind of a pain
They do have the CDN proxy too. Not sure when it was implemented though.
It's a similar process to Cloudflare. Point the NS to them and enable the proxy for a domain or subdomain.
You can also create a standalone pull zone and point your existing DNS provider to the CNAME they give you.
(don't use CNAME flattening with DNS-routed CDNs like Bunny though, if you must use an apex domain then use the CDNs integrated nameservers)
When I tried it last year, their edge compute infra was just not there yet. It could not do any meaningful server-side rendering because of code size, compute and JS standard constraints.
Has this situation changed?
Not sure what you mean with ssr for a CDN?
Edge computing. Cloudflare workers for example.
Bunny has a similarity concept: https://bunny.net/edge-scripting/
I have been using them for over a year. THey have the same flow as Cloudflare, point domain to thier CDN, set CDN Pull Zone to target your server. I havent had to do anything.
They even support websockets.
Why they cant do is the TUnnel stuff, or at least fake it. I have ipv6 servers, and I can't have the IPv4 Bunny traffic go to the ipv6 only sources.
I have used multiple s3 and cdn replacements, and bunny is my favorite. Excited to see a database product in the mix.
we have 3 orange clouds now
Why couldn't they just use SQLite, and not libSQL?
It seems Bunny is competing with Cloudflare. They offer very similar services including CDN, video streaming, databases etc.
Looks cool, I need an alternative to my supabase set-up for little web tools, so i'll check it out!
Is this good for write heavy loads or does it face the same constraints as regular SQLite?
why this over turso or litestream + read replicas?
Bun alert!
> Not every project needs Postgres, and that’s okay. Sometimes you just want a simple, reliable database that you can spin up quickly and build on, without worrying it’ll hit your wallet like an EC2.
Isn't the operational burden of SQLite the main selling point over Postgres (not one I subscribe to, but that's neither here nor there)? If it's managed, why do I care if it's SQLite or Postgres? If anything, I would expect Postgres to be the friendlier option, since you won't have to worry about eventually discovering that you actually need some feature even if you don't need it at the start of your project. Maybe there are projects that implement SQLite on top of Postgres so you can gradually migrate away from SQLite if you need Postgres features eventually?
Marek here from bunny.net. We’re not saying SQLite is universally better than Postgres. The trade-off we’re optimizing for is cost model and operational simplicity.
Even as a managed service, Postgres DBaaS still tends to push users into capacity planning, instance tiers, and paying for idle headroom. Using a SQLite-compatible engine lets us offer a truly usage-based model with affordable read replication and minimal idle costs.
Why choose this over Cloudflare D1?
Disclaimer: I have not used either product; I have used a number of Cloudflare's (mostly free tier) offerings.
In addition to the other points brought up, it looks like pricing strongly favors Bunny once you're outside of Cloudflare's free tier.
Per billion rows read: Bunny $0.30, Cloudflare $1.00 (first 25B/month free)
Per million rows written: Bunny $0.30, Cloudflare $1.00 (first 50M/month free)
Per GB stored: Bunny $0.10/region, Cloudflare $0.75 (5GB free)
Bunny also has a lot better region selection, 41 available vs. Cloudflare's 6 (see https://developers.cloudflare.com/d1/configuration/data-loca...). Even though Bunny charges storage per region used where Cloudflare doesn't, Bunny still comes out cheaper with 7 regions selected. Bunny lets you choose how many and which regions to replicate across; Cloudflare's region replication is an on/off toggle that is in beta and requires you to use "the new Sessions API" (I don't know what this entails).
The main reason I haven't tried out D1 is that it locks you into using Workers to access the database. Bunny says they have an HTTP API.
I plan to stick with VPSes for compute and storage, but I do like seeing someone (other than Amazon) challenge Cloudflare on their huge array of fun toys for devs to play with.
For one they're EU-based, which may be a selling point if you're inclined to divest from US tech when possible.
Cloudflare IPs might not work in Spain during football matches :)
It looks like there might be issues in Italy too.
Good question, the cloudflare speedtest has a bunny in it.
Some ISPs have bad peering with Cloudflare (e.g. Deutsche Telekom). Not Cloudflares fault but it makes it a bad choice if your customers are in Germany.
And Cloudflare is an american company.
Not a technical reason, but given Cloudflare's recent business practices where they hold you hostage if you don't upgrade to an enterprise plan are a pretty good reason to avoid imo.
This sounds a lot like https://turso.tech/ ? Unless I misunderstand, they're both pitching SQLite-for-the-cloud.
Yes, they mention they use libsql. Don't know why I should use them and not the product by the actual libsql authors.
Some European companies migrate their dependencies from US clouds to European ones. Turso is registered in Delaware. Bunny HQ is in Slovenia. Different data related policies apply.
An advantage is the integration with other Bunny services like containers etc.
Also, not sure about now, but historically Turso didn't have to best uptime.