0 reply
0 recast
0 reaction
3 replies
0 recast
3 reactions
1 reply
0 recast
2 reactions
1 reply
0 recast
2 reactions
1 reply
0 recast
1 reaction
1 reply
0 recast
1 reaction
1 reply
0 recast
1 reaction
1 reply
0 recast
1 reaction

im not gonna talk you out of it, i love crazy homelab projects, and i am happy to invest in a lil hardware if it means i learn something new
one thing i have learned after years of trial and error, however, is to segregate my workloads by their demand profile. i have a bunch of "dumb lil homelab" stuff, it runs from 3x thinkcentre nodes in a k3s cluster and a single nvme disk. i still keep a few selfhost services on a remote VPS - can't afford for my password manager to crash just because i update homeassistant. and i use the DL380 exclusively for blockchain stuff and domestic heavylifting, which thrashes disks and crashes offten...
for me, uptime + reliability need to be a key focus in my homelab projects - if i crash my recipes app / jellyfin / home assistant, due to tinkering with blockchain stuff, it blocks my grocery shopping / daily lighting automations / watching TV, until i have time to go be a sysadmin again ... i can't afford that, i want to enjoy these services, not be beholden to them! 0 reply
0 recast
1 reaction