Fox was accidentally put on our work TVs today (usually CNN which isn’t much better). For some reason I was able to see past the makeup and I’m pretty much every host they have now is just the Crypt Keeper underneath 20 pounds of the stuff.
It’s always DNS
My wife mentioned that Teachers are no longer listed as professionals either. They literally want or country to be as dumb as fucking rocks, slave our life away, and die horribly.
It does take quite a bit of upkeep, especially if you don’t use it frequently. I recently found my instance broke due to a bad addon, and then Authelia also broke because NC decided thier OIDC addon is not supported on the latest v32. I was able to re-enable without issue, but still flagged as unsupported.
Sounds like I’m talking myself into Immich already haha.
The issue for me is that Nextcloud has these features as well with App add-ons. I have yet to try Immich because what’s more important for me is the actual backup\upload of my photos than actually browsing them. Maybe someday, but my self hosting initiatives typically involve me chasing a shiny red ball of a deployment, and Immich just isn’t shiny enough for me yet.
Just buy a domain for 10-20 a year and host a dynamic IP updater internally. Just another layer to self hosting and getting off cloud services entirely.
At least you can still program those remotes though. Mine is still going strong after many years.
This is the one I went with along with a supermicro server board. The company has been great as I’ve already needed replacement rack screws and a new control board due to my own foolishness. They shipped me replacements at no charge very promptly.
Same recommendation here. I went through two QNAP units before being fed up and building my own 12 Bay for about 1200. My first QNAP died shortly after the 3 year warranty expired and the second died shortly before. I was able to RMA the second and sell it to recoup some money towards building my own TrueNAS system that I can now fix myself and not rely on proprietary anything.
I was a little unfair in my post towards Proxmox. It really is a great solution and I can’t really complain, but it sucks in comparison to ESX where many “custom” items are still hidden in the cli or custom configuration items,. Many of these things are available in the GUI in ESX which is a pretty rough translation for some that have worked in ESX for many years like myself. ESX isn’t without it’s CLI moments but they are rarely ever needed, and if needed only for drastic measures.
The UI is not very intuitive and really looks quite dated too. ESX, Nutanix and XCP-NG have much better interfaces imo, and if Proxmox could throw some of that extra money they’ve earned from the VMware exodus in their UI it would be worthwhile.
Again, I shouldn’t complain but as I get older there’s not much “tinkering” time anymore, and the less time I have to sift through forum posts or official documentation on why something isn’t working as intended, the more easily frustrated I get.
Don’t go Podman. When I started years ago I installed Fedora with the “containerization” option. This installs podman, not docker as I’m sure most know. I did not.
Podman works great for the most part, but it’s slight differences from docker will have you fighting tooth and nail for certain services to work correctly. And not many (if any at all) have any documentation on getting their containers working with Podman of they don’t start. If you make a GitHub issue asking why or how to get things running in Podman because their docker stack doesn’t work flawlessly like it will in docker, good luck getting help (Mailcow comes to mind specifically here).
Looking back, this decision really shoehorned some very fundamental ideals about containers in my mind, but it was a long fought road I would not choose again. The knowledge I gained about containers with docker would have come soon enough on the easy road.
And yes, you can install Docker on Fedora, but I was much too far down the Podman track before finding out. My environment has changed drastically as of late and most things have been migrated to docker apps in Truenas now, living directly next to their storage as intended (the arr stacks really take a performance hit running their databases over NFS once you have a lot of media for example).
Quick note about Proxmox after coming from ESX myself - it sucks compared to ESX. I’ve tried to move away from it and Nutanix was the closest I could find to ESX, but after my server started complaining it’s drives were not compatible I jumped ship to avoid any write damage to them. I’m downsizing my lab now, I have proxmox running in 3 small NUCs with CEPH storage share and it’s working pretty good. Would love to run ESX or Nutanix instead, but they require a loaf of bread in resource requirements where proxmox only needs a slice of bread in comparison.
As of the last update released on August 1st, the “old” VMs are now visible again. The latest Electric Eel chain also merged all Core features into Scale, so the jump should not be as drastic any longer. I’ve always lived on Scale, but I assume you could try backing up your config and spinning up a new Scale VM and restoring the backup to it. No matter how you dice it though, it will be spicy!
What type of disk (HDD or SSD) and how many disks in the pool?
RAIDZ1 configuration will bring your write speed down some due to data having to write to multiple disks at a time. This is true for most any RAID. Once written, your read speeds should remain the same or improve a bit though.
Have you modified the default unbound config at all? This sounds like increasing the cache size limits and timeframes in the unbound config could help.
I’m actually chasing an issue I’ve always had where everything works great in my environment, but on mobile certain domains take ages to finally load up for me. I think it’s a combination of my Pihole blocking and the amount of domains tied to a page (advertisements and tracking), but would love to figure it out. I work around it right now by flipping wifi off and on again in those instances.
Instead of port 53, I need to run unbound on 5335 (or another obscure port).I believe I also had to make some host level changed for DNS to operate correctly for incoming requests.
Here’s my podman run commands. These might have changed a bit with Pihole v6, but should still be ok AFAIK.
#PiHole1 Deployment/Upgrade Script podman run -d --name pihole -p 53:53/tcp -p 53:53/udp -p 8080:80/tcp --hostname pihole --cap-add=CAP_AUDIT_WRITE -e FTLCONF_REPLY_ADDR4=192.168.0.201 -e PIHOLE_DNS_=“192.168.0.201#5335;192.168.0.202#5335” -e TZ=“America/New York” -e WEBPASSWORD=" MyPassword" -v /var/pihole/pihole1:/etc/pihole -v /var/pihole/pihole1/piholedns/:/etc/dnsmasq.d --restart=unless-stopped --label=“io.containers.autoupdate=registry” docker.io/pihole/pihole:latest
#UnBound1 Deployment/Upgrade Script podman run -d --name unbound -v /var/pihole/pihole1/unbound:/opt/unbound/etc/unbound/ -v /var/pihole/pihole1/unbound/unbound.log:/var/log/unbound/unbound.log -v /var/pihole/pihole1/unbound/root.hints:/opt/unbound/etc/unbound/root.hints -v /var/pihole/pihole1/unbound/a-records.conf:/opt/unbound/etc/unbound/a-records.conf -p 5335:5335/tcp -p 5335:5335/udp --restart=unless-stopped --label=“io.containers.autoupdate=registry” docker.io/mvance/unbound:latest
I just went through my setup to verify dnssec settings in unbound to troubleshoot strange latency when removing random names while browsing. Did you verify the unbound certificate file was created and had the proper permissions? There are also a couple other configuration items in unbound related to dnssec that can be tweaked to improve the implementation.
I do exactly the same thing for all three of these services! My implementation is on podman rather than docker, but basically the same deal.
Apologies, you mentioned specifically your network drive. Interesting article but they give several work arounds for containers that may require host mode, and it appears the non Plex pass image is one of them to resolve this specific issue.
I would go around them and go directly to the source of categorization. It looks like this is the Symantec categorization website in case it’s different from what you’re workplace provides - https://sitereview.bluecoat.com/#/
Welcoming the incoming dowvotes for correcting your comment just like the many similar comments and posts I’ve seen on Reddit, but this is purely a configuration issue.
Transcoding on local network is allowed without a subscription. If you are running your own DNS server (like pihole or unbound) you need to configure an internal “plex.direct” record. You also need to uncheck an option to “treat your WAN IP as internal” option which corrects double NAT issues.
I have yet to see a need to move away from Plex. I paid for the cheap lifetime sub over a decade ago at this point and everyone I invite to my server has no complaints and has not had to pay Plex a dime. I don’t use their plex.tv proxy, I direct connect to my own IP and leave their remote proxy option off in the server and everything works great.
I will check out Jellyfin at some point if Plex makes things more difficult in time, but for now these articles are literally just rage bait in the homelab ecosystem. They enacted this back in April of 2025 already!