It started "indexing" around 9pm. It is now 9am and still hasnt finished reading all the files. Pretty sure its around 10TB or so from last time I ran du. File count would probably be in the millions somewhere because there's a healthy mix of documents, photos, videos.

My photography drive backup is currently running - ETA to completion 15 days at 50MBps.

Initial seed was always going to suck. Should be fast to keep up to date though.

[–] DiaDeLosMuertos@aussie.zone 0 points 2 months ago (1 children)

Jesus what have you got in there ? Like, every file on the internet that has ever existed ?

That's nuts.

Thank you for making today's random thread by the way.

[–] trk@aussie.zone 0 points 2 months ago

By homelab standards, I am a very small fish. I just prefer not to delete stuff.

My Dell setup is all second hand enterprise gear I bought just before the price rises. Dual Xeons and 512GB of RAM which burns a lot of power and has approximately 800% more computing power than I need. But it is fun to watch the blinking lights.

My other setup is made up from old PC hardware I had laying around. Basically just collected the case that best suited lots of hard disks, the highest rated PSU, the best matching sticks of RAM etc... Probably should have cleaned the case out properly in hindsight. Because its all SATA / SSD / NVME it actually draws basically nothing. The spinny drives all shut down when not in use - which is most of the time. The SSD cache has more than enough space to keep up with daily activities.

[–] cdzero@lemmy.ml 0 points 2 months ago (1 children)

That first one sucks on the big batches. I had a similar thing setting up a new 20TB backup drive recently.

I just use rsync to do manual backups every week or so and its pretty quick after the initial monster job.

[–] trk@aussie.zone 0 points 2 months ago (1 children)

I've been using rsync and cron as a backup for like... 20+ years? Always had n+1 hard disks, with one accessed read only by a script that writes a copy to another drive once a day. I've got files from circa-2000 still floating around so the system has been working...

But figured it was time for a deduplicating setup since my data is blowing out a bit. After some trial and error have opted for Duplicacy. Back up running on my home server, and pushing the chunks to my work server (read: server at my workshop as opposed to "work server" as such). Went with a 20:2 erasure coding for that par2-esque bitrot protection.

Somewhat worried since using a program to maintain archives means you rely on that program existing 5ever, and not corrupting anything.... but I figure unraid with 2 parity drives is pretty resilient in terms of hardware failure, and I'm already using Crashplan for remote backups so I've got a fallback.

[–] cdzero@lemmy.ml 0 points 2 months ago (1 children)

You're about 15 years ahead of me.

[–] trk@aussie.zone 0 points 2 months ago (1 children)

Oh yeah, I'm old as shit.

RIP me 💀

[–] cdzero@lemmy.ml 0 points 2 months ago

I didn't mean it as a negative.