Show HN: Minimalist App for Backups to Any Storage
blobbackup.com
blobbackup.com
Obviously this is needed to make block level deduplication possible, but it also means that for this "end to end encrypted" product someone can still tell whether or not you own a given file. This is probably fine for most people but worth calling out.
EDIT: I should make note of that on the docs page really. It currently gives the impression that I'm just straight up calculating the sha of the content.
If, for any reason, you'd like a free test account, just email info@rsync.net.
Thank you for building this!
I figured borg is stable and does the job but others, including restic have problems like memory usage concerns and I'm just a single person shop and don't have any huge assets to backup and already seeing restic/duplicacy choke consuming GBs of memory.
What are the tools that are performant and stable?
I'd like to at least use another implementation besides borg in case a bug corrupts the data.
Encryption, retention policies, incremental backup without periodical full backup and acceptable performance (that people don't fill GitHub issues with performance problems) are requirements and deduplication is a plus.
It used to be per user but now it's per machine and it's even asking you to pay $50/year per machine even for cheap instances that are $60/year and when you have 20 small machines to look after, the pricing looks a bit too unfair.
"The requested URL /resources/regulatory/pricing.html was not found on this server."
The UI is basically a 1-to-1 copy of an older version:
https://blobbackup.com/images/logs.png
http://bvckup.com/img/shot-6.png
"Good artists create, great artists copy", but that's a bit too close for comfort.
Outside of the main window though, lots of other things are quite different and completely independently arrived at. Bvckup2 doesn't have third party storage support so everything else requires a fundamentally different UI to work in BlobBackup. Look at the settings windows and the menus for example.
Similar to how desktop word processors and browsers all look the same; some patterns just make sense for a specific type of software.
Given that this one uses secondary control labels that are verbatim of Bvckup's, has the same layout and opening screen, and nearly identical look and feel, this is certainly not an independent effort.
[1]: https://rclone.org/overview/, [2]: https://rclone.org/crypt/
1. Compression/deduplication. Your data will (generally speaking) take up about half the space it did before even including the multiple file versions.
2. Automation. You won't have to manually do anything after you configure a schedule once.
3. Management of multiple backups.
4. A really nice native GUI
5. All the advantages that backup with file versioning has over sync (which is what it looks like you have). Safe from ransomware (you'll have an older version always even if your infected ones get synced accidentally), projection from human error by being able to revert to older versions, etc
[1] Benchmark on open data set. https://blobbackup.com/10_gb_benchmark.php
Which version of Duplicati was used in this benchmark?
The largest backup a user has reported using this with has been 5 tb and it ran without a hitch.
I'm pretty sure it was the latest duplicati for the benchmark. Maybe the second latest. I forget.
Website does not inspire trust as there is no mention at all about who the seller is.
Terms of use, privacy policy are also relevant, especially in this context where software gets access to all of customer's data...
First time software seller here so I'm a little slow on some things but rest assured, I'm gonna have a privacy policy soon. The EULA is there when you download the software.
BlobBackup doesn't store any user data anywhere other than the user's computer and their backup storage though (and this is encrypted).
https://borgbackup.readthedocs.io/en/stable/changes.html
The downside of borg is that the target is only sftp and not directly to any of S3 and some other targets but there are online services that accept sftp.
On the other hand, restic is new and while this can sync towards many targets, the use of memory gets crazy after a while that my machine choked out of memory and these kind of stability problem is something not mentioned in the docs until you use it.
I've tried several others like duplicacy but this also has a memory problem and both restic and duplicacy seem to have trouble fixing it quickly.
When it comes to backup, I'd like to use more than 1 implementation toward different remote locations in case 1 of the implementation has corruption bugs to render the backup useless.
Trying out kopia to see if it does any better.
Also asuran seems promising but so far too early to test it out.
This used to be (as in, from 2008 to 2018) difficult, one of my past companies made a good deal of money by offering an sftp gateway to S3. But AWS continues to do a great job commoditizing their complements, and tackled SFTP back in 2018 when they launched...
“... AWS Transfer for SFTP, a fully-managed, highly-available SFTP service. You simply create a server, set up user accounts, and associate the server with one or more Amazon Simple Storage Service (S3) buckets. You have fine-grained control over user identity, permissions, and keys. You can create users within Transfer for SFTP, or you can make use of an existing identity provider. You can also use IAM policies to control the level of access granted to each user. You can also make use of your existing DNS name and SSH public keys, making it easy for you to migrate to Transfer for SFTP. Your customers and your partners will continue to connect and to make transfers as usual, with no changes to their existing workflows.”
“You have full access to the underlying S3 buckets and you can make use of many different S3 features including lifecycle policies, multiple storage classes, several options for server-side encryption, versioning, and so forth.”
https://aws.amazon.com/blogs/aws/new-aws-transfer-for-sftp-f...
Because this did not exist for a decade, it seems most folks don’t know it exists today.
Just to turn SFTP on at AWS, you need to pay $0.3/h ($216/mo) and it even costs to upload not just for download at $0.04/GB. If you upload 100GB/day, that's also about $120/mo and if you're storing about 1TB you're paying like $350/mo for such little storage you get.
Who wants to pay for upload? At this point, why don't you just run EBS backed EC2 yourself?
With that kind of pricing, you'd get like 15TB of space at rsync.net and it doesn't leech users for bandwidth and on every parameters as AWS does. (If you use the borg discount, you get 40TB+ at $0.008/GB!)
And at what benefit does it bring for the price difference? I see no point in using it. Maybe people don't care to mention than not knowing it.
I would have used restic directly as a backend for BlobBackup had it not been for the awful prune operation. Downloads a ton of data everytime, prone to errors, and has to repack blobs. Compression would be nice too but they have nice dedupe so it would have been workable without.
What tech stack is this application built in?
Its built in a combination of C and python. Using zstandard for compression and openssl for crypto. Qt for the GUI.
Great product though! Always happy to see projects like this
With object storage like s3, b2, azure, wasabi etc, this just isn't a problem since you can effectively have as many objects in one "directory" as you want.
With local, nas or hard drive storage, you'd be limited by the number of files per directory the file system allows (4.2 billion for NTFS for example). So you're probably fine here too.
Google drive (which isn't recommended for large backups with BlobBackup) has much stricter restrictions on files per directory. So really it depends on the backend but for most use cases it isn't a problem.
BlobBackup has been tested on millions of files backed up to NTFS and object storage yes.
What is your use case for this?
Do you simply mean is it possible to restore backups to local? Or a reverse backup operation from cloud to local?
I've been thinking about this for quite a while, and one reason I've been unhappy with any sort of cloud backup solution is that they're usually too slow for laptops, unless I leave them awake while I'm sleeping.
An ideal backup solution for me would be laptop > NAS (or desktop) > cloud. A backup over the local network should be magnitudes faster, and the always-on NAS can handle a slower, versioned backup to cloud storage.
I know some users run backups to their NAS every 6 hours and then run a weekly backup of their NAS to the cloud using whatever NAS software it came with (alternatively you could sync it through your laptop).