Backups Aren't Simple

> “There are two types of people: those who have suffered a catastrophic loss of data, and those who will.”

When I was a teenager, I was the reason for data loss for my dad, twice. Both times it was because I was re-partitioning a hard drive to install linux.

You would think that taught me a lesson about backups, instead it reminds me every now and again to be grateful for an awesome dad and aspire to handle situations with my kid similarly :)

There are four times in my life I have suffered regrettable data loss incidents.

The first was when the telephone pole outside our house was struck directly by lightning. Not only was it the loudest thing I have ever heard, the current surged through the telephone line, into the internal fax modem, and fries everything within its vicinity. I was 10. I did have backuos, but only only floppy and they didn't cover everything.

The second was storing data in OneDrive - a change to their terms surrounding "lifetime" unlikely noted storage, combined with a client that was unusably slow to download and a deadline for data retrieval meant that I lost most of my files.

The third was SD card failure in digital camera on holiday, the controller chip died catastrophically, leaving the card completely unrecognised. It was a brand new Sony 128GB card, manufactured by Toshiba, and it seemed to be a common issue. I now shoot to two cards simultaneously.

And the fourth time was ... Performing a backup. An errant script deleted the source content, but I'd also deleted the existing backup to free up space for the new backup. I've been weary of using rewritable media for some time now as a consequence, but I think backups themselves are high risk activities.

A friend of mine used to work at Veritas[0] making enterprise data retention solutions. When I spoke about their product as being "making backups", he corrected me by saying:

  We are not in the backup business.  We are in the restoration
  business.

0 - https://en.wikipedia.org/wiki/Backup_Exec

I really like ZFS snapshots with offsite pull-mode sync using Jim Salter's sanoid/syncoid [1]. ZFS is the base for all OS/filesystems on top of it. If you have a good system for organizing ZFS datasets, and separating ephemeral from persistent data (e.g. [2]), then this is 90% of the backup requirements already fullfilled.

    [1]: https://github.com/jimsalterjrs/sanoid
    [2]: https://du.nkel.dev/blog/2026-05-16_rootless_docker_virtiofs_proxmox/

Bit rot was mentioned in the article, but then no solution was offered. I also haven't found a good solution, unfortunately. If you only have a couple of Terabytes you could create check sums of everything and then verify regularly, but I feel this approach doesn't scale very well if you have lots of data.

I’m setting up 3-2-1-ish backups for my infra of 3 hosts, and definitely leaning towards Restic + Backrest.

All my hosts run the same CoreOS setup (https://github.com/ebrahim37/infra-template), where container volumes are placed in one central volumes/ folder and that is the only thing I have to backup.

I plan to implement it like this:

  vps1:
  - restic container with custom sh entrypoint that will backup volumes/ to homelab every 24 hours
  homelab:
  - backrest container, to back up volumes/, do prune/check, replicate repo to offsite
  - rest-server container, will store backups from vps1, homelab, offsite
  offsite:
  - restic container, backs up volumes/ to homelab every 24 hours
  - rest-server container, store copy of backups from homelab

Only caveat is backing up databases, will either have to do: stop container, backup volume/database-data, start container; or use pg dump etc.

The deduplication is nice, you can have a snapshot for each week of the past year without crazy storage cost

If you don't control your data size, it will end up controlling you and your backup choices, which will eventually lead to many avoidable, disastrous outcomes. Somehow, this gets missed in data storage and backup planning.

"Also make sure not to run backups at 2AM or 3AM, or things may get scary."

I never though of daylight savings times causing such an issue. Learned something today.

"encrypted, chunk-level deduplicated, GFS-rotated, point-in-time archived, cloud, 3-2-1 backup solution" is now my newest password, no commas. (Don't tell anyone!)

I agree with this sentiment! For our Abejora timesheet SaaS, setting up backups was one of the harder parts. We had to change direction a few times while implementing it.

We finally got a nice setup with CloudNativePG + Barman. This allows for point-in-time restores, but there were a lot of lessons to learn along the way.

- The various types of (database) backups (logical, binary, onsite, offsite, snapshots, write-ahead log...) in combination with the various types of data (database, files, cluster configuration...)

- In our earlier approaches, we tried to preserve the old database volume if it was not corrupt, and use that in our restore. This caused so many complications, because you are fighting the recommended approach. So now, when we need to restore, we always restore from backups and the 'live volume' is dropped.

- For a restore, we just spin up a completely new Kubernetes cluster, instead of trying to restore in-cluster. This is a lot easier.

- Many object stores allow for retention periods, which you can put to good use to prevent malicious or accidental removal of backups. HOWEVER, not all of them are really 'locked'. In some services, you can still override the lock with a forced delete; in others, you can still remove the project holding the storage buckets, which will delete the buckets, and so on... so test those things, instead of just blindly depending on a 'retention period' claim.

- We now automatically run a scheduled restore with verifications on a weekly basis. This requirement does shape your environment, so keep that in mind! There is also the question of how you can reliably and automatically verify that the restore restored the latest data (of a live prod environment). Various solutions exist here, but most are not very elegant!

Honestly, this is only worth it if you are already handling sufficient volume. If you are just starting out, then the easier approach is to just go with a hosted database, which will handle backups and point-in-time restores for you.

> There are two types of people: those who have suffered a catastrophic loss of data, and those who will.

This is an empirical claim, but is it grounded in reality?

99% of people don't backup, and that's probably the right choice because the risk is low, they can't meaningfully improve their restoration rate themselves and they don't care enough about their data to classify its loss as catastrophic.

When a person hears someone talk about the importance of backups, but also haven't heard friends/family suffer this date, they will rightfully ignore this warning.

Or do we all follow the best practices as it relates to backups, exercise, sleep, nutrition, accounting, house maintenance, ... Etc.

I do not. Backups aren't near the top of that list.

This stuff just isn't very important to most people and that's okay.

On my home PC I use restic, and (shockingly) I run it manually (via a script). I have a terminal banner that nags me when backup hasn't run in 24 hours, or displays spanish verb tables when it has (so I will notice if the banner doesn't show). I've been burned in the past losing a backup cron, and dead man's snitch seemed overkill for my needs here. And I've also forgotten what backup system was in place on a prior install, and it took me way longer than I liked figuring out how it was setup when I needed to do a restore.

Obviously for production systems automation & monitoring is the way. I kind of like this setup on my own machine though.

For my small library photos, keepass passwords, etc around 30gb, I use syncthing, which will backup on my pc, on my parents pc (they live in another country), on my sisters pc and on my phone. They also use this triangular mode to backup their stuff. Syncthing is setup to only sync through tailscale or home routers network.

I've given out jwz's link to several people:

https://www.jwz.org/doc/backups.html

(maybe copy/paste this into browser instead of following link, since referrer from hn apparently does something)

My setup: rsync to Synology NAS, snapshotting there, sync to remote Synology NAS over VPN. (The other NAS is at my parent's and they have similar setup syncing to me)

The best part: it is just files, not some proprietary archive. I have high confidence in this. I can restore with the most rudimentary tools.

This setup has worked for me well on linux, windows and now mac. I migrated to new computer&os by restoring the backup mostly.

For my Oracle Linux systems, I rely on the vendor ReaR RPM package for the baseline backup. This is similar to the "Ignite" utility that I used in my HP-UX days.

https://relax-and-recover.org/

This is two-fold useful, first as it enables boot from backup media for total recovery, second in that it creates a backup.tar.gz that holds everything that was not explitictly ignored in the /etc/rear/local.conf file.

I'm running Oracle databases on these systems (and the local.conf is configured to ignore the datafile directories). I have standby databases (not Dataguard in that I don't have online redolog replication) that allow me to recover them to the primary in a disaster, with some loss of committed transactions.

I finally have rsync configured for some scratch temporary files.

I have dallied with btrfs snapshot replication for home directories (in a loopback mount). My vendor support (via a CSI number) has been on and off, so I don't use btrfs in areas that we need it most.

My replacements do not like the complexity (and I am retiring).

One of our customers decided to go for off site backups with us - they had been a customer for 15 odd years for other services. Backups are not our core business, its one of the services and I got around to setting them up sooner rather than later.

Two weeks later they had a fire in one of their huge printing machines which caused quite a lot of direct damage due to heat and rendered a lot of equipment broken due to smoke/fumes. It was also, as it turns out, a week before the first COVID 19 lockdown in the UK.

Their backups landed on a XFS file system with reflinks enabled. I cloned their repo and fired up NFS with their VMs running on our gear. It took me another hour to spin up another OpenVPN server (CA etc) for them to use and sort out a few other details (coffee doesn't brew itself).

They ran via VPN out of our data centre for most of the pandemic.

That was an utter triumph but I also have some rather less triumphant stories about backups and lack thereof. Let's skip over those 8)

I'm polishing my backup system right now, and found that `tar | zstd | gpg` plus a small vibe-coded wrapper is pretty much all I need.

GNU tar has its own incremental index via `--listed-incremental=FILE`. Unlike Borg or Restic, which have their own more complicated repository formats, this leaves me with just one additional file (the `.snar`) alongside a dumb, portable full-disk tar archive.

The nice part is that, unlike repositories that require both read/write access patterns, tar can compute deltas using only the small `.snar` file, while the main `full-disk.tar` can remain buried in write-only Glacier storage. This makes it a much better fit for Glacier's write-once model and 180-day minimum retention.

My current plan is to upload a full-disk tar to an external HDD + Glacier on a weekly cadence, and the `.snar`-based deltas daily. That gives me a pretty simple cloud backup solution for a few bucks a month (after burning through $100+ of free credits).

A bit of a tangent, but the linked rsnapshot tool is about 7k lines of perl code. The COPYING file is 20 years old and it's still maintained, the last change was 6 months ago. I'm feeling really old.

[1]: https://github.com/rsnapshot/rsnapshot/blob/master/rsnapshot...

They are simple.

The issue is companies love to fire QA people who should be the ones testing these.

Microsoft says we don’t need no SDETs, our products can ship full of issues.

What are the hostages, I mean customers going to do about it?

I have no sympathy for any major corp who suffers data loss. Now individuals who lost grandmas wedding photos, that sucks.

But my recovery strategy for that is simple. After scanning I emailed it out to one cousin who emailed them out to a few more. Plus I have a cloud backup ( I have no idea where the the original files are).

Data is meant to be shared after all

3 rules of backup (don't remember where i read this)

- do several backups (not using the same hardware brand or host if you do it online)

- check your backup (if something was wrong, you have another backup from the first rule)

- do not store your backups at the same place (if your place get flooded or burn, it will be pretty handy)

Something is better than nothing. I accidentally spilled liquid on my macbook just the other day. It's still in the shop. Luckily, I periodically clone my drive with CCC onto a single external SSD. Don't wait to set up the perfect backup solution, just have a copy of your data in more than one place, that's a good start imho.

Huge fan of borg + vorta for backups and time Maschine like „Jump back in time“. Wrote about it years ago: https://simon-frey.com/blog/borgvorta-is-finally-a-usable-ba...

I have a cron script that rsync my entire home directory every day with a remote server. Since rsync doesn't remove remote files, I re-create a backup from scratch the first day of each month and store a copy of the previous one.

Using old standardized open source tools means it survives system updates without having to fix anything for years. I got my laptop stolen/lost 3 times over the past 15 years, and I have always been able to restore everything the next day on a new laptop, in the time it takes to transfer the files over the network.

I haven't given much thought, I'm sure there is a realistic scenario where this strategy would fails but I haven't found it yet.

Ran duplicity - it's really a blunt hammer, lots of duplication, rotation, etc.

Then switched to Restic - so much better - highly recommend this.

concerning backups, when it comes to disaster recovery you need a strategy for restoring as well, sometimes it's not only the data you would need but the systems as well

>Motivated by our success in developing this solution, we try to use it to backup the homelab with its 10 Docker containers. But later we find out from logs on the individual machines that backups are failing. The reason being that many Docker containers like to create root-owned files, and if you’re not careful you can create a cronjob running as the default user.

and a privesc also.

+1 for restic.

ZFS snapshots + restic backups to backblaze for my homeserver. My secret sauce is a healtchecks.io instance that blows up my phone if ZFS scrubs throw any errors, when local snapshots fail, or when restic checks or backups fail.

As an average-techie person, the only piece of data that i truly practice 3-2-1 is photos and videos. 1 on my phone, 1 on my small server running Immich, and 1 is on Google Photo(sadly, the quality is loss, but i can live with that).

Most of other data i only have 2 backup, usually 1 at my local machine(code) and 1 online(github). Yeah it's not ideal, but another question to ask yourself, is it truly worth it. I trust the engineers at MS doing much better job at backing up their data than me.

On the topic of backups, 7zip recently changed its handling of the FILE_ATTRIBUTE_ARCHIVE bit on unzipped files which caused an issue with some old software I was using. This is from the old DOS days with the BACKUP/RESTORE commands - the bit was set if any backup was made of the file. It's comically naive because it only assumes a single backup and this bit was forgotten to the sands of time until it was disturbed by some meddling kids!

Backups isn't real until it's restored

don't be like me folks

Time Machine (to NAS) + iCloud (w/ ADP) + local copy is the lowest friction path I have found.

"Obviously, none of this is worth anything if you don’t actually test restores."

One of my teachers claimed that contrary to common behavior, no one actually valued backups; it was restores that were worth paying for.

I use ZFS and do off site backups using rsync.net.

Yep, backup is a verb, restore is a noun. Testing restores is where the real work begins.

> as long as you run restores every 6 months

uh no, in a company, as a sysadmin, you try your backup files at least every month

Truenas, Crashplan gives me a 3-2-1 with rolling, incremental backups at a pretty low price

Amazon are finding this out atm :-P

https://archive.is/Ay7RJ

I don't want backups. I want synced duplicated data in geographically separated by shared credentials.

So is apparently making readable site.

Software like https://restic.net/ does a good job. Few choices of backend

Accepts stuff from pipe too so you can just pipe mysqldump or pg_dumpall without intermediate files

Very decent options for checking repo integrity, personally (well, at work too, we use it on few hundred user machines and servers) I also added "roll a dice for backup and try to restore it" test script to make sure it is working

Decent deduplication too so "store year's worth of weekly snapshot" is very sensible strategy once you exclude the "constantly changing and useless" (caches etc.) files out of it

can mount FUSE directory with all backups on sensible OSes.

https://kopia.io/ does the same +GUI but a bit worse on CLI front (it insists on keeping local config which makes scripting a bit more involved, but not much)

When it comes to software, it's almost never simple. What people think of as simple is rarely so.

For example:

Requirement:

"Just copy a file from folder A to folder B."

Minimal translation:

"Implement the capability to copy any file, in any format and of any size from SharePoint, located at a configurable path, with appropriate authentication and access control checks, then stream it in chunks to a different, configurable path inside an S3 bucket, also with appropriate authentication and access controls in place. Ensure that any disruption in either service which may occur while the file is streaming can be recovered from at the point it failed instead of having to restart from the beginning. Ensure that the retry mechanism is built-in and that the retry window is configurable; if the file cannot be copied within the specific time window, then an error should be sent via email to a configured email address. Ensure that the entire transfer is encrypted in transit... If the file happens to be a folder, then you must copy across all of its contents recursively up to a certain configured MAX_DEPTH to avoid DoS and ensuring that the system does not get caught in an infinite loop due to symlinks pointing to a parent folder... In this case, send an email to the configured address... Etc... Etc..."

And the thing is; if you tell AI "Just copy a file from folder A to folder B." - It will not meet your 'basic' needs because even if it does a great job at filling the gaps in your requirements, it will still take shortcuts. In order for an AI to avoid taking shortcuts, it would have to make you fill out a questionnaire and make you sign up for and configure services; it would not be a pleasant user experience. The user experience cannot be pleasant, because the AI cannot read your mind and it cannot know your intent.

> So our first principle is to have a backup, i.e. a copy of your files someplace else. So far so good.

So who are we? A household with a localhost administrator?[0]

Then you are surely doesn't need encrypted, chunk-level corporate level bla-bla-bla.

You need a Syncthing copy to some other device not at home - and an additional backup procedure to maintain the history and protect against PEBKAC errors - which can run on your local Syncthing copy or/and the other one.

And no, nobody needs your 555GB of RAWs of sunsets/flowers/precious_family_moments you shot - not even you.[1]

[0] well considering the tone of TFA and most of the comments here

[1] that's Instagram/Google Photos/iCloud for nowadays, totally with "Remember this day N years ago?"

[dead]

[flagged]

[dead]

[dead]