Let’s do another round of this. What’s going on on your servers, any cool things? How’d the recent CVE flood tangent you and did it change how you see/do stuff?
Added no-new-privileges to all my outward facing containers :P
Submitted 22 hours ago by tofu@lemmy.nocturnal.garden to selfhosted@lemmy.world
Let’s do another round of this. What’s going on on your servers, any cool things? How’d the recent CVE flood tangent you and did it change how you see/do stuff?
Added no-new-privileges to all my outward facing containers :P
Feeling better every day having my stuff under my control and away from prying eyes of trackers and other unsavory stuff.
My automatic weekly backup ran for the first time last night. Everything worked nicely.
I practiced restoring my offsite backup this week. Good experience.
Then I printed and laminated the instructions and stored in two locations.
Just curious. How long does lamination last? I assume light hurts paper.
If it’s automated, why only run it weekly?
I’m always super excited for this kind of stuff :D
This week, I’ve mainly been handling the fallout from the PHP vulnerability that occurred in late October. Following that incident, I made a few changes: I’ve opened only the necessary ports to the outside world, and the admin panel is now accessible only via VPN. I’ve created an inventory list for each server, clearly detailing which services are running and their versions, so that the next time a CVE is released, I’ll know immediately which servers are affected. In addition to running daily automatic backups, I periodically restore them to a clean machine to verify they work.
Remember: Always treat your servers as if they might be lost tomorrow, and make sure to have a thorough recovery plan in place.
I recently got around to setting up OIDC for my internal apps using Pocket ID.
Sonarr v5 adds OIDC support, currently only in beta
Jellyfin SSO plugin
Nextcloud with the Nextcloud OIDC app
Mealie, Paperless, and Immich natively support OIDC
Calibre-Web I migrated to AutoCaliWeb, which supports generic OIDC
Sadly Navidrome doesn’t natively support OIDC and forward auth is a bit jank due to the subsonic API
Just completed my set of needed services by adding nextcloud to the list. Now life is good. Jellyfin: check. Navidrome: check. Pihole: check. Invidious: check. Nextcloud: check. What else? lol
You’ll find more stuff for problems you didn’t know you have. :D
Upgrading one of my powerhouse Dell R410 proxmox nodes to 128GB of smoking hot PC10500 DDR3 ECC Ram from 64.
Now it’s cutting edge.
I got new rack gear. I don’t 3d print so I ordered metal face plates for my 10 inch rack. got them here https://www.rackmountplates.com/
I am not affiliated. just a happy customer that stumbled on a solution to a problem. Sharing in case others have similar needs
Looks pretty cool, quite pricey though compared to 3D printing. I’m glad my friend printed mine
I’m into this as a regular thing!
Pinned, I think a lot of us really enjoy seeing what others have been doing to take back and make improvements to our own setup.
Glad you like it! I tried to go for a 2-3 week cycle as I realized every week is a bit too often (less comments). I also tried to do it alternating with the community on slrpnk but they’re currently down. I often forget to do these anyways :D
Anyone can make these, of course. I always enjoy doing them and talking about stuff I enjoy.
After the recent plex player debacle I’m once again trying to come up with a way for me and my diaspora to share media using Jellyfin, but without relying on Tailscale. We have several media libraries, and some of the library hosters are not too tech savvy, so coming up with a solution to replace the ease-of-use of plex hasn’t been easy.
Hey, I did the same. It’s not as bad as it seems. Get yourself a domain name from cloudflare and use it. I got mine for $10 a year. Then just installed cloudflare on my Debian server and all that (I’m sure you know the rest) then made users on my admin jellyfin dashboard and started sending people their usernames and passwords. The only difference for them is now they would need to enter the server URL on the first page of whatever they use then username and password. Good thing these jellyfin clients have the of entering the server in your face with their “connect to server” button. Let me know if you have any questions.
Careful with using CF Tunnels for streaming content. It’s against their ToS to use the free tier for streaming content, so they can technically turn it off whenever they want from my understanding.
Procrastination. Want to move may lab off my pi onto my mini PC.
I have done the backup off the pi, just need to restore it but motivation doesn’t line up when the family is out of the house.
I bought some hardware last winter… still sitting in boxes. I do have adhd though
Part of it is I want to shuffle around some stuff, like switch my mobo out on my main pc and use its mobo in the new server
I have proxmox installed on the mini PC with a Debian VM ready to go, just the step of moving everything is the problem.
i have an asustor nas and i set it up with my hdd raid 1 drives as volume1, and installed my ssd later as volume2. now docker is constantly making my hdds tick and I can't move it to the SSD because asustor don't allow it ☠️ so I'm looking at reeeeaaallly testing my backups, wiping the device and reinstalling everything. It's all in docker so it's theoretically easy enough 🤞
OMG patching all the vulns. I actually pulled my wordpress offline overnight for that one. Still waiting on the Ubuntu patch for that IPv4 kernel CVE.
That reminds me - It’s time to do No NAT November again. This CVE gives me an excuse. Last year, the Nintendo Switch Switch and Steam were the big standouts. I think the Switch 2 will actually work with it now.
Oh, and my workplace is shifting offices, so my offiste backup is not offsite. Must take care to not burn the place down. I just put new fans in my PoE switch and now I can hear a server fan is dying too. There’s a lot of maintenance work to all this.
Finally took the leap and wiped my NUC to start doing things over. Originally had a family member who works in IT set everything up, but he did the absolute bare minimum and I screwed some stuff up along the way the first time around, so I figured now was as good at time as any to teach myself to set things up from scratch!
I also have moved from Portainer, to Dokploy, to Arcane for docker management, and I actually really like Arcane. I’ve also accomplished something I’m really damn proud of, which is getting HTTPS working on Tailscale (so, for example, arcane.(my subdomain).ts.net but with HTTPS!) alongside Pocket ID for all of it. Maybe it’s just me, but I had trouble finding clear documentation on how to do this, no matter how much I searched. But I documented it for myself, at least. And it’s probably that I lack a lot of knowledge in this field, too.
I have some questions, and although I know there’s a chance this gets buried, I figured I’d try asking anyway. Little skittish about making a thread myself.
My NUC is running Debian 13, and I was wondering some security things. Everything is behind Tailscale and I don’t plan on exposing anything to the internet (even through Tailscale Funnel, I’m using Tailscale Serve for the HTTPS.)
Thank you for reading my very long update! And any answers to my questions are very appreciated. Hope everyone else is doing well!
F2B/FW: If it’s only available locally/via VPN, I think your threat model is your LAN. People in your wifi or rogue smart home stuff. I think a firewall makes more sense here, but if it’s just 1 machine, you should be fine with ufw or the like. If you have several VMs that are supposed to be separated, the solution would depend on the virtualization software.
Unattended upgrades are way better than none at all, both for system and docker. Of course it depends on how important it is that your services are “always” available and I get the NAS struggle though.
Hey, thank you for the comment!
That makes sense, that I would be tightening security on the LAN. It’s not just one machine, I have a Raspberry Pi 4 as well running a flavor of Debian (and the Synology NAS), so I would assume a firewall on both the Raspberry Pi and my NUC would make sense? I don’t have any VMs that I’m aware of, and I’ve heard of VLANs but don’t know anything about those.
Services don’t always have to be available! It’s just me and my partner who are accessing them, so nothing like a lot of people who need media playback. I’ll have to weigh if it’s worth fighting with the NAS over auto reboots, but I’m leaning towards yes as an experiment. I’ll have to actually look into how to do unattended Debian security updates, though.
I am trying to find affordable 2x16Gb ram modules that are just right to run on my asrock J5040 and push my proxmox to more possibilities.
Because my needs for ram are increasing since its not just about have the services availble anymore. I plan to add gitlab and k3s and with the rest of my services the 16GB ram are pretty tight.
Good luck! I’ve been watching auctions for some weeks on ebay when I needed 2x32GB SODIMM some months ago and eventually got a pair for 230€. Fucking expensive but cheaper than most other offers I found.
This is a realy good price nowadays. Even more when it was DDR5. I only need DDR4 but atm its round about 80€ for one. Makes 160€… Fir this old stuff. And it even feels like a good price. Pretty sad somehow xD
Got my hands on an old laptop. Since I started using nix a little while ago, and want to migrate to NixOS for my servers, this will be the starting point. Started cobbling together some code to commission it as a smart tv platform. KDE Plasma BigScreen is what I’m looking at atm.
Holy shit are you me? I started doing exactly this with an old laptop, NixOS and Plasma-Bigscreen too!
I’m currently in the process of writing yet another script. This time my aim is to make one that will setup a server from a fresh Operating System install.
I want to do this for two reasons. The first is so that I have an alternative to backups. It would also help with major Operating System updates because sometimes those are a bit messy.
The second reason is because I want to help other people setup a simple, minimal and more secure setup for a home server as well.
It will all be POSIX portable too so it won’t be limited by extra requirements. I’ve already made a lot of progress with it because I started doing the same with my personal computer. I’ve been able to focus on that because my home server is super simple and I’ve been able to ignore all the recent vulnerabilities.
Don’t create another Omarchy please!
I have a similar thing but I’m using Ansible. That way, I don’t have to handle all the error checking etc myself.
I don’t like bloat or AI/LLMs and I focus more on minimalism to reduce surface attack areas while still focusing on security. There’s nothing flashy about how I plan to approach this project. It’s going to look ugly.
My main focus is to help create communication for local communities and not create tech that enables control over other peoples freedom.
I highly doubt my project will ever leave my community and that’s fine, I’m not interested in making a name in an already crowded space.
Not quite server related, but in the middle of a workday this week, I decided I was done with Windows on my home PC. Installed Zorin OS and things have been going really well. Any issues I’ve had have been really minor and mostly revolve around me being picky about theming.
I haven’t found a native mp3 tagger that makes sense to me yet, so I just installed Mp3tag. I’m open to suggestions on that front.
On the server side, it’s mostly just keeping on. A friend has me considering Home Assistant, I just don’t know if the value is there for me. I need to do some housekeeping today and update and clean up unused containers.
I have opinions about music tagging! I’m picky; I don’t trust any amount of automation to find the exact right release with the right album art I want, so I use GUIs instead of a CLI -
MusicBrainz has Picard, and that’s pretty slick. Also includes a handy renaming tool for keeping files organized. Quod Libet is also worth a look. I’m not as familiar with its feature set, but it’ll run on a toaster.
One day I might try to tackle Beets. Heard lots of good about that as an auto-tagging solution, so maybe check that out, too.
Yeah, same. Earlier this year I thought “Hey, people seem to love auto tagging with MB, Picard, Beets. I should try it.”
But that was short lived. I don’t want release information outside of artists, titles, albums, year. I just have no need for it. I know how to get what I want from mp3tag.
I am looking at using Beets to keep things clean, maybe find missing artwork, and fetch lyrics. But the vital information I’ll keep reviewing.
I haven’t ever gotten comfortable with how Picard works, but I’ll check out QL. Thanks!
Working on a custom rust server / client for self hosted obsidian lifesync. Goal is to be significantly faster than existing solutions as well as being hassle free.
Apart from that I finally figured out why tailscale wasn’t working on my pfsense and that enabled me to move the majority of my services to be behind the VPN.
I hosted Donetick. It was awful. Everything is processed server-side, which makes it crazy slow. Data should be backed up locally on device. Shame because it seems an otherwise sophisticated software.
Sounds great, I’ll try it out.
I’ve just uninstalled Super Productivity because all the data was client side and server sync was problematic. I’m always going to use it on the home network. This sounds like what I wanted with time tracking and everything.
Other than this… I’ve been trying to keep up with keeping everything up to date with all the vulnerabilities with PHP, Debian Kernel vulnerability, etc.
Oh I also set up Homarr. I didn’t think I needed a dashboard. But this has been amazing.
I hope it works out well for you. I don't have problem with server-side processing, I just wish it would be backed up locally so I didn't have to stare at a spinning circle for 5 seconds just to mark a task as completed. Currently I use Tasks (tasks.org) and there's a "completed" option right in the notification.
Using nix to maintain my fleet with automated updates so I didn’t really need to worry about it thankfully. All my services are nice and up to date yay
What I’m excited for today is to progress personal slop project (be gentle) I started yesterday. Hoping it’ll just be one or two weekends.
I’ve been hallucinating a transcoding suite/workbench. Includes experimental sampling matrix of ffmpeg settings with a composable / layer-based project configuration; rust egui+mpv desktop client. Then the hosted orchestration server can distribute ffmpeg jobs across my worker machines (via my local headscale tailnet). Output produces a project+metadata bundle that includes hand-picked source:result JXL frame comparisons alongside both human readable reports and client readable configuration that makes the projects (more or less) reproducible (with worker tool env versioning via nix flake.lock) and logs provenance.
I intend for this to be a replacement to the piles of crappy python scripts I’ve been using for the last decade; my current workflow is to move files into host-labeled ingest folders with a an argument string in a text file then ssh to the host to run the job.
But what I’m most excited for is the experimentation matrix. I hope that it will reduce my electricity usage and disk thrashing; it will let me sample settings for ~1minute of footage around the sample frame and let me iterate my tuning much faster. Currently, it’s faster (in terms of hands-on time) to just transcode whole episodes or even entire seasons and multiple times with different settings and then just discard the ones that underperform. Extremely wasteful; assembling Matrix is something I’ve always wanted, but never had the time to build (and still don’t, hence slop).
I was 429ing a little too liberally, so I upped my burst rate. This week, just like last week, most of my traffic was bots. But if a real user does come by, I want his or her experience to be pleasant enough.
I shut everything down only to discover nothing runs on php, while shit was going down at work. And I just recently discovered how to run docker compose updates through Komodo which is great.
I love Komodo! Have you attached Renovate as well?
I might check that out but for now I just added the existing docker-compose file as stacks in Komodo and enabled the auto pull and auto update flags
I’ve taken the opportunity to update most of my hosted services to newer image versions.
They’re all behind private access (Tailscale, mTLS, IP whitelist etc.) so I’m not usually too worried. I update images manually whenever I can find time on a slow weekend, so this wasn’t strictly speaking related to the recent vulnerabilities. Volume data and container definitions are backed up automatically once a day and I also take manual copies of the old images when I update them.
I’ve switched to new major versions of Immich and Jellyfin in the process and I’m very pleased with the new releases, amazing great work from the developers, they’ve been killing it lately.
I’ve been tinkering with some custom images that I use for various things like syncing specialized data (logs off FTP services, archive email over IMAP from my mail provider, grab Garmin data copies, run monitors and log parameters to Influx, run port forwarders etc.)
I typically want to lock down this kind of images as much as possible, so I tweak them towards running with an unprivileged user, remove all caps and only grant strictly necessary ones, make the storage read-only inside the container etc.
Some needs have been recurring inside this type of containers, like having a decent process manager to keep some things running and restart them, or having a decent cron service. So I’ve been toying with the idea of making a “universal” image for this kind of thing and maybe publishing it for others to use as a base too. If I get around to it I’ll be announcing it here.
I deleted a bunch of services I’m not actually using on my home server. Nothing used php, but it was good to do some house cleaning.
I also started a blog (Hugo is much nicer than I thought). The procrastination has helped me get a lot of other things done.
I’ve been a Joplin user and using their Joplin Sync server for a few years now. Last night, I migrated everything over to Obisdian and just finished setting up a self hosted obsidian sync. I will say that the self hosted sync was nothing short of a puzzle for me but finally nailed that down between my phone and desktop app. So, I have the same as Joplin now for that functionality.
What got me started on this little side journey, is I’ve been seeing more and more pretty cool uses for Obsidian and have a lot of great things about it. I’m not dumping Joplin yet, but am giving this a solid try before I make any hasty decisions.
Do you have any experience with logseq? I am using it as I looked for an obsidian alternative that is FOSS. I don’t really use much of its possibilities though, so I don’t really know how they compare.
I went from Joplin to Logseq (now Logseq OG)
Joplin just didn’t click for me, but Logseq saved my life when I took on a new role at work.
I’ve not switched from the markdown to the db version, but after 3 years of almost daily notes it’s ok, but it needs a good spec machine when you’re searching across all the files
No, but I used this as part of my basis for decision:
noteapps.info/apps/compare?note_app=joplin%2Blogs…
You can select which notes apps you want to compare features against.
From my logs nothing bad happened to any of my services. I turned off my entire homelab for a few days until every single app/service had an update. Then I turned everything on, updated and that was that.
One cool thing I have is github.com/lklynet/hypermind totally pointless, but I like it.
Recently picked up a lenovo system x 3650 m5. Came with 16 GB DDR4 which is already worth more than I paid for it. Now to find the time to mess with it…
I’ve been focused so much on work-related stuff involving DST changes that I neglected my own systems!
ComradeMiao@lemmy.world 4 minutes ago
Thanks for this post, something I miss from earlier selfhosting communities.
I have simply running two SFF Lenovo Thinkcentres—one for all local services and another for all online services. They could run on one but I keep them separate for security and stability reasons. I often crash the local one by running out of space…
I finally got to building a second backup recently which I am happy about.
Honestly, since moving and downsizing from several services at home and at work, this solution works. I run out of space frequently but it forces me to not hoard things I will not actually use. My previous rack server seeded something like 50tb as I kept everything I leached to benefit the community, now less possible.
This has also made me move away from services that are not needed.
[ ] I would like to offload all my home videos to an external drive as I am running out of space for actual services and Jellyfin media.
[ ] I would like to consolidate all of my local and online databases rather than run a new one for each container stack.