

You can try Colota as a way of recording your phone location with its own filtering before it uploads to Reitti/Dawarich


You can try Colota as a way of recording your phone location with its own filtering before it uploads to Reitti/Dawarich


I have my important server stuff (docker and immich etc) duped via syncthing to an old external USB drive attached to my office mac which then has Backblaze Personal Backup running. That backs it all up with no storage limit for £70/year on their 2 year plan.
The only downside is if the drive isn’t connected for a few days, it checks it all again on reconnection.
I have about 4TB of files most of the time.
This way I have 3 copies of my files: 2 local, 1 remote.
EDIT: forgot that I also have 1yr retention on the files so in theory I can grab anything that I deleted up to 1yr ago.


So this has been interesting to me as I’m travelling at the moment and wireguard has been blocked on the guest wifi I’ve been on across the hotels, I’ve had to rely on cellular instead.
I’ve just tried Teleport on my Unifi router and that works. I believe it uses wireguard, and it is taking up to 20secs to connect, but I’m now curious as to what it’s doing to bypass the VPN restrictions that are blocking plain wireguard.
I had Infomaniak for a while, and tried the kdrive, as like you, it was the cheapest cloud drive storage. However I gave up with Infomaniak as in the end they felt a bit, er, fussy in the way they ran it all. It just didn’t fill me with confidence.
For me, for all archiving, the simplest solution has always been Backblaze Personal backup. Yes its an extra cost per month, but it will also backup all of your main computer, and any external drives. You can add an encryption key too. The only downside is it being a US company.
I have a 5TB external USB that dupes all my docker stuff as well from my server to my Mac, so it’s an all-in-one backup solution for me.


deleted by creator


deleted by creator


I use a Pangolin reverse proxy with OIDC (PocketID) for family access to services, along with CrowdSec. For the Immich app access which needs to bypass auth login through the reverse proxy, I use ‘link share’ in Pangolin that gives me header tokens that can be entered in to the Immich app under Advanced settings.
I’ve been an Immich user for over 2 years now, so it’s been a journey for me to implement it to this standard.
Or as someone else suggests, try CloudFlare with something like Google Auth login. Just be aware that you are then exposing all your traffic to Cloudflare. I take that as a small sacrifice for simplicity.


Thanks for that. Read Pt1 and it all makes sense.
I’ve found that Music Assistant has worked well for group casting audio for me with Google Minis and Chromecast Audios. Although I’ve not had it running all day, but it has removed the hassle of it stopping from WiFi disconnects from my phone.
Always looking for new ways to do things, so will look out for pts 2 & 3.


Take from this anything of use to you: I syncthing my important Unraid server stuff to a MacBook running Backblaze with a 14TB external drive. DNS is handled by NextDNS in the cloud anyway, so it’s just Immich and other random family oriented services for me. I use Pangolin for exposed services.
You could use syncthing (or Resilio) over a tailscale or headscale network or whateveritispeopleuse to a remote PC.
My Docker compose files and VMs are backed up once a week via plugins on Unraid, which of course get duped via syncthing to the MacBook.
I did try a Hetzner Storage Box for a while to replace Backblaze, but the hassle of having to keep an eye on the syncing was a pain and I fell back in to Backblaze on the MacBook as it’s a set and forget (as long as they don’t change the filetypes ignored…). I may go back to a Storage Box again.
I do also want to look into redundancy with a spare server though so it becomes quicker/easier to get it all up and running in event of failure. This is where selfhosting starts getting serious!
Just chucking this out there to help with the fediverse.


Hi mxdcodes. I have been using your app for the past month in a sort of testing phase with Darawich. I have used GPS Logger and Reitti since October last year, which is still running, but Colota offers a lot more exciting possibilities of data collection and so I’ll probably switch across pretty soon. I got hit by the bug a while back of it stopping once entering a geofence, but I saw that got fixed promptly so thanks for all of this.
I love all that you and others who code for the (wider) selfhosted community offer, so let me say that for every person criticizing your app and all the effort you have put in, there are hopefully 1000x more of us who are truly grateful for what you and others do. I find it odd that people who are concerned about having their location history hacked by someone are actually wanting to record their location history in the first place. Seems the obvious answer is to… not track your location history with ANY app.
Keep up the good work!
Good to hear!
However I may have spoken too soon with Colota… it stopped registering my location after a first trip, and wouldn’t restart tracking until I restarted my phone (GrapheneOS Pixel 9). I need to look at the logs and maybe raise an issue.
Dawarich has had some great updates in the past few weeks, and I’m hoping Colota + Dawarich is my future family tracker system as they both have a very nice feel to them.
GPS Logger + Reitti is my background system atm while I mess with Colota & Dawarich.
I’ve been using Colota this weekend sending data to a Dawarich server. I like that you can filter out inaccurate GPS readings by setting an accuracy threshold, so this removes any false points that I tend to get with other location apps.
I’ve tried GPS Logger for Reitti, and the new Android Dawarich app, but so far, Colota seems the best (early days). It has the nicest ui as well.
In terms of what to enter, you need to find out what HA requires for the URL to receive data, then set that in the Colota app. Colota will just store the location data until you set a remote end for it to connect to.


I was just thinking this week, that those who self host (and more importantly, those who program the code we self host), are at the front line of the modern digital resistance: in the sense that the world is burning due to the greed of the tech bros that run our daily lives. Convienience for the masses is what gives them power over us, and any one who rejects their systems is helping to fight back.
Voting with your wallet helps, so not giving them your money is the first step. Then managing and keeping your own data private is the next one.


2024 was the year I got more serious with self hosting and migrating away from the cloud offered by Google etc. But 2025 was the year I pushed to run all my own services and get the family on board as well; trying to educate my kids with running our own services (the wife is so not interested!).
There were some really cool projects released last year and some oddly well-timed ones as I was looking for various services, and Jotty was one of those!
Thanks so much for you work and rest assured amongst the negativity you may receive in certain corners of the web, there are people truly appreciative of yours, and others like you, hard work.


I had Nextcloud running for several years (VM is the best way IMO, I would avoid the Docker AIO). However I found Filebrowser and it rocks as a file share service. Filebrowser Quantum is a fork with more feature as the original no longer has a maintainer. The most I’ve had someone upload to it was 300GB.


I use Filebrowser Quantum if you are happy opening up a port for it. It supports 2fa. Also requires Docker which isn’t too difficult.


We also do the sauté onions (which is just onions cooked slooooowly). They caramelise and become sweet, add some generic chicken seasoning to them (I use a salt/paprika mix from the general store), tinned baked beans in tomato sauce, rice, and that’s all. Spice it up with some jar jalapeños and its a damn fine meal for nearly no prep or cost.
Yeah, Caddy was working fine, but the issue was me tinkering with it meant having to reload Caddy for the updated config to work, and that would break any connections people were using for file transfers etc. Also, it isn’t as quick for reverse proxying file transfers.
Therefore trying to run private and public services through it was limiting when I was also trying to tweak it constantly for my homelab.
I’ve found Traefik to be better in that it auto reloads the config live as you edit it, and it’s been faster for file transfers on my 1Gbps fibre.
And now I’ve split my services to separate public/private reverse proxies, that takes the pressure of having to keep one proxy always live. Pangolin uses Traefik, and so do I for my direct services through my firewall, and that makes life easier when only dealing with one type of proxy service.
I too am using a Cloudflare tunnel for my public facing services (such as WordPress), and that also allows you to put the WP login page behind another auth login as well which is great for security, so I do also vouch for Cloudflare.
I’m using Pangolin for private services on a VPS.
Plus, I have one service that is direct to my home IP for file sharing to one particular remote IP that is the only service directly through my firewall.
Therefore I have 3 ways my services are accessed and this has been the game changer for me recently, as previously I tried to run all this through one Caddy reverse proxy directly to my router and it gets painfully fragile mixing public/private services through one bottleneck when you’re tinkering as a selfhoster. So splitting it up has helped massively.
Good tip with the Cloudflare alts though!
Not about certs, but I found the opposite moving from Caddy to Traefik: in amending the caddyfile, I had to restart the Caddy service which would kill any open connections that were being used. In Traefik, it’s dynamic, so can update the services without having to restart Traefik itself.
My main move to Traefik was due to file transfers for reverse proxy services like File Browser slowing down after a period of time, whereas in Traefik, they stayed at a constant speed.
I now run Traefik as part of Pangolin, so most config is set by that.