I use Syncthing for this type of task on my PC and Phone and it stores a copy of the shared folder on the server with the option for file versioning. Having a Server is optional by the way.
AFAIK, Syncthing clones the entire folder across peers (the server is just another peer it seems), which isn’t ideal for my use case Do you know any current way to configure it for selective syncing?
I don’t think it can do selective syncing. I’ve been also searching for a similar solution but didn’t find one. Finally opted for syncthing with my most important files. Other files I can get them via web using filestash.
Owncloud supports selective sync, and seems a lot better for performance compared to Nextcloud.
Alternatively you could roll your own with rclone which is essentially an open source alternative to mountain duck. Then you can just use a simple connection via SFTP, FTP, WebDAV, etc…
Non-OCIS Owncloud still needs a dedicated database and recommend against SQLite in prod
I’ve looked at rclone mounting with the --vfs-cache-* flags. But I’m not sure how it can smart sync like mountain duck or handle conflicts elegantly like the Nextcloud/Owncloud clients do. Let me know how to set it up that way if possible
This! Im planning on getting this set up on a spare pi one of these days™. You get a free premium acc on the tracking sites, so you can track where tf all those planes and helicopters above your house are going
I did this for a hot second (already have RTL-SDR set here) but the current location of the RPi is just bad for reception and moving it closer to some window would mean connecting through wifi (can’t lay ethernet cables, renting) and that’s bad for other services where low response times are preffered/needed (pihole) :(
Instead of connecting it to WiFi, have a look into power line adapters. They route your internet through the copper wiring in your house.
I have a router in my subterranean ground floor linked to a power line adapter, a wired router in my front room a floor up so my PC, TV, Playstation, etc are connected via LAN, and another power line 2 floors above that plugged into another WiFi router running in bridge mode, which supplies WiFi to the top two floors, and another playstation wired in to that router
Basically it means that my ground floor router is hooked to the internet and everything else in the house that needs wiring in is wired in because of the power line, and the WiFi is coming from 2 routers, one on the top floor and one on the ground.
My ISP thought a WiFi router on the ground floor of a 4 storey house was a great idea, but they’re stupid. WiFi should be in the highest point of your house.
With a few Power line adapters you can sort your internet out for £25
Already got ssd as a nfs share in my openwrt-based router before that I did have it set up on the rpi. I did want to do offsite backup into that disk originally but I’ve got “only” ~100Mb/s up/down speed here so I didn’t want to risk slow-downs etc (but now that you remind me, borgbackup should be rather light on traffic!).
NEMS being a whole OS is a pitty, I like the possibility to have multiple different services there.But you are absolutely right I could have a offsite resource monitoring for my Hetzner setup with these, thanks!
I definitely do not count it against them as long as they know how to human at the interview. I just review the code as I would any repo.
The only thing is that with regular projects I tend to go “I noticed on your GitHub you have project X that uses technology Y, etc etc”. With H projects I just go “do you have experience with Y” and let him choose how much he wants to share about the project. So far they remain vague on the non technical details and I let them leave with their dignity intact.
So, ranked, way ahead of candidates without visible projects, but slightly behind people with projects we can discuss in detail in front of the people from HR ;)
The spacing in the email screwed up the formatting:
Dear Andre,
I’m Gianpiero Morbello, serving as the Head of IOT and Ecosystem at Haier Europe.
It’s a pleasure to hear from you. We just received your email, and coincidentally, I was in the process of sending you a mail with a similar suggestion.
I want to emphasize Haier Europe’s enthusiasm for supporting initiatives in the open world. Please note that our IOT vision revolves around a three-pillar strategy:
achieving 100% connectivity for our appliances,
opening our IOT infrastructure (we are aligned with Matter and extensively integrating third-party connections through APIs, and looking for any other opportunity it might be interesting),
and the third pillar involves enhancing consumer value through the integration of various appliances and services, as an example we are pretty active in the energy management opening our platform to solution which are coming from energy providers.
Our strategy’s cornerstone is the IOT platform and the HON app, introduced on AWS in 2020 with a focus on Privacy and Security by Design principles. We’re delighted that our HON connected appliances and solutions have been well-received so the number of connected active consumers is growing day after day, with high level of satisfaction proven by the high rates we receive in the App stores.
Prioritizing the efficiency of HON functions when making AWS calls has been crucial, particularly in light of the notable increase in active users mentioned above. This focus enables us to effectively control costs.
Recently, we’ve observed a substantial increase in AWS calls attributed to your plugin, prompting the communication you previously received as standard protocol for our company, but as mentioned earlier, we are committed to transparency and keenly interested in collaborating with you not only to optimize your plugin in alignment with our cost control objectives, but also to cooperate in better serving your community.
I propose scheduling a call involving our IOT Technology department to address the issue comprehensively and respond to any questions both parties may have.
Hope to hear back from you soon.
Best regards
Gianpiero Morbello Head of Brand & IOT Haier Europe
I’ve been using it for about a month, and love it.
My one complaint: self-signed certs on reverse proxies seem to break the android app backup. I’m not sure why, but internal CA seems to make things angry. Its more likely to be a local setup issue than anything in immich, but frustrating to pin down.
My current backup strategy is BTSync, which while super easy to get going is a pain in the ass to look up old images. Using direct IP on the app works perfectly, and the DNS lookup only works internally anyways.
All that to say that I’m probably going to use it and remove the btsync approach in a couple months.
I also have internal only traffic, but I still use let’s encrypt. I self signed for a couple of years, but switching to proper certificates made things much simpler and better. Especially on mobile.
I use a combination of my own domain and caddy. and duckdns, since my domain registrar does not have an api caddy can use, but I can point my domain to my duckdns domain and it works 👍
Yeah I think I do too in the attic somewhere. Mid chips on those things where a bitch back then when it first started up. I think they got better though.
They made a software jailbreak?!?! So I just looked this up and I have Splinter Cell and MechAssault. I may have to dig that thing out and give this a try.
I was aware of forgejo back when I first started hosting Gitea. Didn’t see much of a diff back then so I just went with arguably more popular option at that time.
About few months after it’s mostly just because I’m too lazy of a person.
Forgejo is a fork of Gitea. As of now I don’t think they have diverged much. So they’re (still) about the same. It was mainly created because of the takeover of the domain and trademark by a for profit company. Not because of different functionality.
As explained in this discussion this seems to be a problem with the web interface only, caused by the framework used by the interface (Svelte). It seems that getting subpaths to work with Svelte is not supported, and the Immich devs are probably right to think it should be fixed by Svelte, not by Immich.
Sorry that’s on me. That’s one thing I removed from the pastebin file on purpose, it’s here in the real one. I copy paste my path to be sure it’s correct.
If you’re running it on a separate server you would have to expose the docker API to it, which you might not want to do. It might be easier to just run it on the same server and proxy it from the bps.
I haven’t researched this, but my gut tells me one should be able to connect the two servers via Wireguard (direct tunnel, tailscale, zerotier, what have you) and discretely access the docker api without making it publicly available.
The company says services like Plex, Pluto TV, Sling TV, Starz, and ZDF will introduce support later this year.
I always had the impression that plex was really slow when it comes to implementing new features. I’m definitely looking forward to a chromecast alternative though. Being locked into googles DNS gives me problems due to it prohibiting streaming from my local server via hostname.
I really hope it will be implemented in Jellyfin as well since I can’t get my installation to work with Chromecast at all (most likely again due to the DNS issue)
I solved that by adding an 8.8.8.8 ip to my pihole interface. Because of how TCP/IP works, this has the fewest hops and is, therefore, the one to be used. I’m blocking all outbound DNS traffic for good measure.
By adding do you mean blocking it in pi-hole or somehow redirecting it to your pi-hole dns server?
I currently have it blocked in my router and can confirm this by trying to ping 8.8.8.8 without any response. If you mean redirecting to your pi-hole I would really like to know how to do it
I didn’t add it to any lists, but to the network interface itself. You know the output of ip a? The one pihole listens on (wg0 in my case, because wireguard) has something like, say, 10.0.0.1, but also8.8.8.8. So when a DNS packet is spit out by chromecast to go to 8.8.8.8 UDP port 53 - my pihole happily answers that request. You could also do a separate unbound instance on a new virtual interface with a quad8 ip and just forward everything to pihole, if you fancy.
selfhosted
Active
This magazine is from a federated server and may be incomplete. Browse more on the original instance.