A well built power supply will have a current limit on the connector to prevent the cable melting problem, but that means if you have a card that needs 200w, but only power it via 1 connector you are going to run into problems
A well built power supply will have a current limit on the connector to prevent the cable melting problem, but that means if you have a card that needs 200w, but only power it via 1 connector you are going to run into problems
Keycloak to provide OIDC, although in hindsight I should have gone with Authelia Authentik
There are very few things more obnoxious than an asshole with unsolicited parenting advice
https://www.servethehome.com/everything-homelab-node-goes-1u-rackmount-qotom-intel-review/ would probably be a better bet for a router
I moved just about everything to Route53 for registration - I run my own DNS so I don’t need to pay for that, and it’s ~40% cheaper than Gandi for better service.
Now I just need to move my .nz domain (R53 supports .{co,net,org}.nz, but not .nz itself?) and the 2 .xyz domains that are “premium” for some reason so R53 won’t touch
For anything that is related to my backup scheme, it’s printed out hard copy, put in an envelope in a fire safe in my house. I can tell you from experience there is nothing more stressful than “oh fuck I need my backups but the key to unlock the backups is in the backups fuck fuck fuck”.
And for future reference, anyone thinking about breaking into my house to get access to my backups just DM me, I’m sure we can come to an arrangement that’s less hassle for both of us
I was in the same place as you a few years ago - I liked swarm, and was a bit intimidated by kubernetes - so I’d encourage you to take a stab at kubernetes. Everything you like about swam kubernetes does better, and tools like k3s make it super simple to get set up. There _is& a learning curve, but I’d say it’s worth it. Swarm is more or less a dead end tech at this point, and there are a lot more resources about kubernetes out there.
They are, but I think the question was more “does the increased speed of an SSD make a practical difference in user experience for immich specifically”
I suspect that the biggest difference would be running the Postgres DB on an SSD where the fast random access is going to make queries significantly faster (unless you have enough ram that Postgres can keep the entire DB in memory where it makes less of a difference).
Putting the actual image storage on SSD might improve latency slightly, but your hard drive is probably already faster than your internet connection so unless you’ve got lots of concurrent users or other things accessing the hard drive a bunch it’ll probably be fast enough.
These are all Reckons without data to back it up, so maybe do some testing
Pretty much - I try and time it so the dumps happen ~an hour before restic runs, but it’s not super critical
pg_dumpall
on a schedule, then restic to backup the dumps. I’m running Zalando Postgres in kubernetes so scheduled tasks and intercontainer networking is a bit simpler, but should be able to run a sidecar container in your compose file
If you figure it out, I know several companies that would be more than willing to drop 7 figures a year to license the tech from you
We would like to apologise for any unpleasant experiences or doubts caused by the miscommunication of these terms
We are sorry we got called out
Yeah, they are mostly designed for classification and inference tasks; given a piece of input data, decide which of these categories it belongs to - the sort of things you are going to want to do in near real time, where it isn’t really practical to ship off to a data centre somewhere for processing.
Seems pretty reasonable. At the end of the day people have to eat, so projects like this either trundle on as hobby-and-spare-time projects for a few years until people get bored and burnt out, or you find a way to make working on the project a paid gig for the core people
This is an “x-y question” - what are you actually trying to achieve?
Clearly you are concerned about… someone… knowing your home IP address - who, and why?
As in, hardware RAID is a terrible idea and should never be used. Ever.
With hardware RAID, you are moving your single point of failure from your drive to your RAID controller - when the controller fails, and they fail more often then you would expect - you are fucked, your data is gone, nice try, play again some time. In theory you could swap the controller out, but in practice it’s a coin flip if that will actually work unless you can find exactly the same model controller with exactly the same firmware manufactured in the same production line while the moon was in the same phase and even then your odds are still only 2 in 3.
Do yourself a favour, look at an external disk shelf/DAS/drive enclosure that connects over SAS and do RAID in software. Hardware RAID made sense when CPUs were hewn from granite and had clock rates measures in tens of megahertz so offloading things to dedicated silicon made things faster, but that’s not been the case this century.
It’s not just let’s encrypt - the common names of any SSL cert issued by a public CA have to be recorded in a public certificate transparency log. You can use tools like https://crt.sh to search the logs
Previously Gandi, but they’ve jacked up their prices and cut features, so in the process of moving to AWS Route53.
My main requirements are:
I’d considered doing something similar at some point but couldn’t quite figure out what the likely behaviour was if the workers lost connection back to the control plane. I guess containers keep running, but does kubelet restart failed containers without a controller to tell it to do so? Obviously connections to pods on other machines will fail if there is no connectivity between machines, but I’m also guessing connections between pods on the same machine will be an issue if the machine can’t reach coredns?
500w TDP Vs 360w on previous generation - cooling this thing is going to be fun