Is syncthing for production?

Can I generally use syncthing for production? How much can I consider it stable if just sharing files between one device with many other hundred devices(one folder for each device and one main sharing a folder with all the others)?

1 Like

Version starts with a 0, so it’s not super stable, and we break things every now and then,but I suspect some people use it in production.

1 Like

Uuu ok! So for now I will use Amazon S3 and https://transfer.sh . Dreaming to use Syncthing in the short future! When do you think that the 1.x stable release will be published?

Whenever its ready and settled what it is. I was told irssi took 18 years for that to happen.

3 Likes

You can see here what the maximum size of share and network of computers people are using syncthing for (or, at least those who have opted to send such statistics)

https://data.syncthing.net

someone shares ~24tb of data, someone shares between 200 pc’s, some server has 500gb ram and some other server (maybe same server) has 72 cores. That probably counts as “using in production”.

6 Likes

Yeah! you convinced me! I was just not good interpreting the stats cause the 100% is not the sum of 95% plus 5% and I still can’t understand why atthe moment :slight_smile:

oh! because it’s a statistical distribution not a sum. Those are “percentiles”. 100 is top most (whatever the highest of that number is). 50 is “half of all values for this thing fall at or below this number”. et cetera.

3 Likes

Thanks! I created the Centiles page in wikipedia. I would suggest to show all the 100 percentiles and not only 4 of them. Anyway from the percentile 85%, it looks that at least 1000 devices are usign syncthing for production: at least 220.000 fiels shared for each device at least 500gb of data storage shared, webs of at least 10 computers sharing files, at least 8 cpu core.

1000 professional users is a number that many startups (financed huge coins) in this sector are not even dreaming about today…

I’m not sure I’d use the number of files / data as an indicator of how many users are using Syncthing in a professional setting.

Who’s to say that the people syncing lots of files are the ones using Syncthing in a professional setting? I can imagine lots of companies using Syncthing to sync a small number of files, and lots of independent users using Syncthing to sync a lot of data (people tend to have very large music collections, for instance).

1 Like

If some people share so much many songs, maybe they are professionists about the music? anyway the usage is similar to be a professional suage. Even the average emule or bittorrent users is rarely managing sa so huge quantity of files I suppose.

Professional means that you get paid for what you do. Someone who has lots of music isn’t automatically a professional.

I think you’re drawing very large conclusions from no data. There’s no way to judge who’s using Syncthing in a professional capacity, and who isn’t.

ok, professional mean that, production mean something else. It mean that many people can use a software without big troubles, and that they can use it for work even if payed directly or not form that usage.

1 Like

If you look at the usage graph, you see this spiky trend, which I suspect mimics monday - friday behaviour, which I suspect is most common for workplaces.

1 Like

I agree with canton7 here.

My “personal” part of syncthing use is quite enormous: >3TB of data in >300K of files.

I have few computers in few locations that I use, and with synthing I keep them all in sync, in a way that i can work from any machine at any time and everything is there. The huge numbers come from huge photo/music collections and their management software DBs. And a huge document archive with all kinds of personal/not personal stuff which is actually the smaller portion of those numbers (although an important one).

In terms of reliability it’s decent now. I had some bad data (almost) loss incident about a year ago, but it was a combination of issues, not only syncthing’s fault, luckily i got off-site backups…

For now it’s deeply integrated in my workflow, and I hope it’ll go on like that :slight_smile: However, stuff breaks here and there, and sometimes manual intervention needed to fix things.

As mentioned it’s v.0.x… so no big promises :wink:

Thanks! From my side I will just use Syncthing to send data from a server into the folders of multiple clients, then the clients will manage these data by themselves, moving out the copied file. So I should don’t have so many issuses about data deletions… and if deleted for mistake, I will be able to regenerate htem if needed too. Syncthing is super laso to deliver the same content to multipledevices, and I will use it for this reason too.

By “manual intervention” i mean physical/remote access to the client PC itself.

For example: “a critical bug introduced that prevents syncing and self-updating”. In that case you’ll need someone “who knows stuff” to access the affected machine to apply the fix.

Things like that are extremely rare as we go forward, but have happened…

1 Like

Thanks! I am already using remote access to all the devices. I suppose that it would be very hard to break the linux remote access using Syncthing…

1 Like

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.