Finally mustered the executive function to clean and replace the faulty fan on the secondary hypervisor that has been sitting offline for the past month or two.

I sacrificed 10Gbit connectivity so that I could install those two enterprise SSDs that were gifted - those should come in real handy! I may be able to reconfigure the system differently in the future, but I’m doing the best I can with what I have given the state of storage and RAM prices.

Updated the UEFI firmware, currently waiting on a full memtest to run before I start doing things again ‘cause the update blew away all my configured settings. [eye twitch]

Once I get the Sharkey instance moved back over to this secondary hypervisor, I can get busy with replacing those shitty low-end Samsung SSDs in the main server.

What do you mean

I have been SPOILED, holy shit

@maddy datacenter SSDs are so good…

@maddy dont look up what those cost now!

@privateger Getting those new SSDs for my main server was bad enough - I think I’ll stay ignorant.

Bleh, that downtime was more than I would have liked. Sorry about that, fellow instance users.

At least it’s done, and everything looks to be running properly. I can poke at the main server when I have some energy, and without interrupting the instance!

Downtime was two minutes longer than it should have been since I didn’t realize the RJ45 port I plugged between the server and the switch wasn’t configured as a trunk port.

Thankfully I realized what was going on pretty quickly and got it configured as a trunk port as soon as I could get into the switch config.

@maddy I swear, the hobbyists are often better at keeping their stuff running than the big dollar shops are... 💜

@anyia It certainly depends, but on average, I’d agree.

@maddy lmao my server has 9% used apparently at 4,151 hours

@niko Oof! This server’s main SSDs are both cache-less WD SN770 1TB drives. About 15,000 hours and 8% wearout.

What model of SSD(s) in yours?

@maddy WD SN770 1TB as well lol

@niko Huh, must be the kind of workloads you have running on there.
I’m certainly not using my hardware as efficiently as I could be.

@maddy my desktop on the other hand has 11,015 power on hours and 3% used with a SN850X 2TB

@niko Hell yeah, I have a few SN850X between some different machines, and those things are solid!

@privateger @maddy just don't leave them without power for long (they're allowed to start losing data fast, within a few weeks iirc)

@maddy i think i've just had the flash too full for too long so it's been using the remaining blocks too much

i've been working on freeing up flash space slowly

@niko Mmm, yeah, that might do it.

@noisytoot @privateger Pffahahaha, I don’t think It’ll ever be offline longer than a day if I can help it!

@maddy hm. my newer server's SSD is at 60% for 36k hours, and older one is at 100% for 48k hours. that tracks I guess.

@maddy don't worry the replacement for the older server is sitting next to it, I just need to do the thing

@izzy The hardest part, sometimes.

@maddy frfrfr
I should probably check the writes on the new server's SSD too I guess

@maddy actually I need to swap the uhhhh spare optane drive I have here into the current in use new server because the one I swapped in was beat to shit

@izzy Oh noooo

@maddy

Filesystem                    Size  Used Avail Use% Mounted on
/dev/mapper/rootfs-root       899G  758G  137G  85% /

​

@maddy oh jesus postgres is using 20% of my disk

@niko @maddy pastgres