# Towards Syncthing 2.0
**URL:** https://forum.syncthing.net/t/towards-syncthing-2-0/24072
**Category:** Development
**Created:** [March 29, 2025, 12:12pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072 "2025-03-29T12:12:09Z")
**Posts on this page:** 15
**Page:** 4
### Author:  [@xor-gate](https://forum.syncthing.net/u/xor-gate)
#### Post date: [June 12, 2025, 7:38pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/63 "2025-06-12T19:38:37Z")
Hi all, as the maintainer of official Syncthing for macOS wrapper I was wondering if there will be a roadmap towards 2.0, when will v1 for example be phased out and bugfixes not backported. In some communication I did read there are no big breaking changes (yet) the major one is on-disk database switch to Sqlite. But I did read there will be some API changes as well in the future, this can and will break the Syncthing for macOS integration. I will try to catch up with the changelogs so I can maintain also two versions. As long as syncthing v2 will be RC the macOS version will not auto upgrade from v1 to v2 via sparkle. Thanks all for the hard work and make this happening.
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [June 12, 2025, 8:28pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/64 "2025-06-12T20:28:08Z")
For your purposes, you probably just need to worry about command line flags: [syncthing/relnotes/v2.0.md at main · syncthing/syncthing · GitHub](https://github.com/syncthing/syncthing/blob/main/relnotes/v2.0.md)
2.0 is “imminent” after which further 1.x releases will be “as required”…
---
### Author:  [@fulstop](https://forum.syncthing.net/u/fulstop)
#### Post date: [June 13, 2025, 9:39am UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/65 "2025-06-13T09:39:56Z")
Do we have any performance comparisons between v1 and v2 ? Will it be faster ? Need more or less cpu / ram / io / disk space ?
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [June 13, 2025, 9:51am UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/66 "2025-06-13T09:51:47Z")
Almost certainly yes, or no, depending 😀
---
### Author:  [@iaw4](https://forum.syncthing.net/u/iaw4)
#### Post date: [June 13, 2025, 12:21pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/67 "2025-06-13T12:21:34Z")
Wishlist: generally more user-friendly messages when the syncing is stopped or not starting. This may require passing messages forth and back between servers. It would cut down tremendously on the message traffic in this forum, too (“why does it not work?”)
(Reach goal: a way to see what’s on the other computer and/or ssh tunnel access to do it by hand. useful when one computer is behind a firewall.)
---
### Author:  [@terry](https://forum.syncthing.net/u/terry)
#### Post date: [June 13, 2025, 5:25pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/68 "2025-06-13T17:25:14Z")
I would say it’s about the same but the database is triple the size with some of the db 2Gb - 3Gb. Not sure if that size will impact the overall speed, but if the dbs are on HDD, it’s going to be painful
Note, I am an extreme user with lots of devices and folders
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [June 15, 2025, 8:47am UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/69 "2025-06-15T08:47:12Z")
> [@terry](#):
>
> if the dbs are on HDD, it’s going to be painful
Maybe, maybe not? I mean, that was pretty painful in v1 as well, and one reason for the larger database is there are more indexes resulting in (hopefully) more targeted reads instead of scans…
---
### Author:  [@bpavuk](https://forum.syncthing.net/u/bpavuk)
#### Post date: [June 17, 2025, 5:58pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/70 "2025-06-17T17:58:45Z")
> [@calmh](#):
>
> From any real programming language it will probably be nicer to use the auto generated clients based on either the gRPC spec though
Honestly, I was incredibly surprised to see that Syncthing does not have gRPC nor OpenAPI/Swagger spec when I was writing a Ktor-powered Syncthing API client. The resulting code is incredibly messy to say the least. Glad to see gRPC coming!
---
### Author:  [@bpavuk](https://forum.syncthing.net/u/bpavuk)
#### Post date: [June 17, 2025, 6:08pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/71 "2025-06-17T18:08:22Z")
> [@xor-gate](#):
>
> maybe its a good time to (partial) rewrite Syncthing macOS to Swift only instead of Swift+Objective C (bridged).
It is, without any doubt.
> [@xor-gate](#):
>
> Maybe with the help of some AI coding tools.
They’ve proven themselves incredibly useful in translating from Java plus tight coupling to Android Framework to Kotlin and somewhat more platform-agnostic APIs. Maybe, your rewrite will also bring [cross-platform support](https://developer.apple.com/documentation/xcode/configuring-a-multiplatform-app-target) to iOS and iPadOS. Some things such as API client can easily be done in multiplatform way.
---
### Author:  [@xor-gate](https://forum.syncthing.net/u/xor-gate)
#### Post date: [June 18, 2025, 6:43am UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/72 "2025-06-18T06:43:25Z")
Looking good, thanks @calmh for pointing at this
---
### Author:  [@Chris073](https://forum.syncthing.net/u/Chris073)
#### Post date: [July 6, 2025, 4:22pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/73 "2025-07-06T16:22:45Z")
Is there any news on when 2.0 might release? No rush by any means, just curious. Are we talking weeks, months, or years?
---
### Author:  [@marbens](https://forum.syncthing.net/u/marbens)
#### Post date: [July 6, 2025, 5:11pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/74 "2025-07-06T17:11:38Z")
[GitHub milestone](https://github.com/syncthing/syncthing/milestone/119) says July 1st, but I think with [Syncthing v1.30.0 (2025-07-01)](https://forum.syncthing.net/t/syncthing-v1-30-0-2025-07-01/24574), it’s being delayed by some time, my guess is August 5th (next release cycle), but it’s just a guess.
ping @calmh
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [July 6, 2025, 5:44pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/75 "2025-07-06T17:44:38Z")
Something like that. Once there are no egregious bugs… I’ll drop a new rc now with the latest bugfix, and may try get a new log format in there before full release, but it’s definitely a matter of weeks at most.
---
### Author:  [@bt90](https://forum.syncthing.net/u/bt90)
#### Post date: [July 15, 2025, 1:43pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/76 "2025-07-15T13:43:50Z")
Interesting hackernews find:
>
>
> If you’re building Python async apps (FastAPI, background jobs, etc.) with SQLite, you’ll eventually hit two issues\- Opening/closing connections is fast, but not free—overhead adds up under load\- SQLite writes are globally locked\aiosqlitepool is a tiny library that adds connection pooling for any asyncio SQLite driver (like aiosqlite):\- It avoids repeated database connection setup...
While the lib isn’t useful to us, the underlying performance optimization might be.
tl;dr: try to keep sqlite connections around for better performance
---
### Author:  [@calmh](https://forum.syncthing.net/u/calmh)
#### Post date: [July 15, 2025, 4:20pm UTC](https://forum.syncthing.net/t/towards-syncthing-2-0/24072/77 "2025-07-15T16:20:32Z")
Yup. We have that, courtesy of the database/sql connection pooling.
[Previous page](https://forum.syncthing.net/t/towards-syncthing-2-0/24072.md?page=3)