trilobite

I moved from next cloud to syncthing some months back. I had nextcloud as an app for Truenas scale. Several times after app dates, next cloud would stop running and would have to setup up everything again.

Syncthing is OK but 2 things annoy me:

A. I get huge amounts of conflict file generated that use up space

B. File sharing with family is complicated. I tried to setup a share account that everyone uses but as syncthing works with device ids, it refuses two accounts from the same machine. I share my Linux laptop with my wife. We each have our own linux account. I've got syncthing running but can't even get my wife's account to sync because I get errors that device I'd already exists.

I don't want to go back to next cloud just for file sharing. I don't generally like the idea of relying on one service for multiple objectives (calendar, file sharing, etc.).

Is there a way to get syncthing to do what I want?

I've been running VMs on some old DELL T110ii but realise that I've loaded it a bit too much so want to leave it doing the job of NAS with Truenas Scale and move all my VMs to Proxmox. The idea is that I would have two optiplex that provide redundancy. Truenas Scale has got me used to ZFS but clear may not be an option with Optiplex 3020 as ZFS is pointless with one SSD. Has anyone got some similar arrangement and has their VMs and containers running on these simple desktop machines? How are you managing high availability and resilience?

Hi folks, I've got a VM that is running my Firefly iii instance and Paperless instance as containers. A lot of work and time goes into managing these tools and I want to make sure I don't lose them. This is my setup:

Turenas Scale machine 1 -> VM1 - Docker containers. The VM sits on its own dataset in Truenas.

I replicate the dataset to Truenas Scale 2 one a week and this machine only goes on on Sunday to save power.

I Rsync the dataset to a 3rd machine where there is a hard disk that I store offsite.

I recognize that I could lose up to one week of work but that is nothing compared to the human hrs spent building those databases from scratch.

Apart from snapshotting e rsyncing every day, what else could I do to make this more resilient without increasing CAPEX and OPEX costs?